Python in Dr. Dobb's Journal (Nov) 
Author Message
 Python in Dr. Dobb's Journal (Nov)

Sports fans,

I'm told that a review of the book "Programming Python" is scheduled
to appear in the November issue of the magazine "Dr. Dobb's Journal."
The text of the article is already online, at:

    http://www.*-*-*.com/

It's an interesting review.  He likes both the book and Python, but is a
little less optimistic about Python's future prospects than some of us.
Draw your own conclusions, but it seems a bit too early to call the race.
For instance, the camel book was first published in 1991, 5 years before
the first python books; we have awhile to catch up.

Not necessarily picking on Perl,
Mark L.
---
"We're on a mission from God" --The Blues Brothers



Tue, 21 Mar 2000 03:00:00 GMT  
 Python in Dr. Dobb's Journal (Nov)

Quote:

>I'm told that a review of the book "Programming Python" is scheduled
>to appear in the November issue of the magazine "Dr. Dobb's Journal."

        Neat review.  BTW, looking at DDJ's editorial calendar,
February 1998 is "Scripting and Alternative Languages"; is anyone
planning to write an article for it?

        According to DDJ's author guidelines at
http://www.ddj.com/ddj/authors.htm, the lead time for an issue is 4
months, so the deadline is either just past, or is just a month away.
I doubt that's enough time to get something ready, since it says they
require a month just to look at a proposal.

        Other possibly-relevant DDJ issues in '98:
   May 1998 - Numerics and the Year 2000 Problem
   June 1998 - Patterns and Object-Oriented Design
   September 1998 - Communications and Networking
   November 1998 - Distributed Computing

Another topic for IPC6...

        Andrew Kuchling

        http://starship.skyport.net/crew/amk/



Tue, 21 Mar 2000 03:00:00 GMT  
 Python in Dr. Dobb's Journal (Nov)

[DDJ review of Python and "Programming Python"...]
: [...] likes both the book and Python, but is a
: little less optimistic about Python's future prospects than some of us.

The argument the reviewer presents is overly brief. I will take the
liberty of expanding it to show its weakness:

"All the programmers who will ever need to use a scripting language have
already been born and have already learned Perl, Tcl, and VB. The
gain in productivity in choosing Python is not large enough to offset the
cost in learning it."

Absent increases in life-extension, the need for any scripting
language should cease to exist in ~130 years, since the cadre of
scripting language programmers will have all died off. If you subscribe
to the notion that maybe, just maybe, there exists a continual flow
of new {*filter*} into (and new/old {*filter*} out of) the programming profession
and that software tools are evaluated afresh by new programmers (and
sometimes by unsatisfied old programmers) on a semi-continuous basis, then
in the long march of time the tools that are better have a good probability
of gaining ascendency.

P.S. VB is a scripting language??



Tue, 21 Mar 2000 03:00:00 GMT  
 Python in Dr. Dobb's Journal (Nov)

Quote:

> The argument the reviewer presents is overly brief. I will take the
> liberty of expanding it to show its weakness:

> "All the programmers who will ever need to use a scripting language have
> already been born and have already learned Perl, Tcl, and VB. The
> gain in productivity in choosing Python is not large enough to offset the
> cost in learning it."

Which also contains the fallacy that one who has learned a specific
skill can count on ALWAYS being in an environment where that skill
is relevant/useful.  Case in point below.

Quote:

> P.S. VB is a scripting language??

Yep.  In its incarnation as "VBA" ("Visual Basic for Applications"), it
is the built-in scripting language for such products as msExcel and
msWord.  In its incarnation as "VBScript" it is a scripting language
for Internet Exploiter.  It can also be used to drive any w32 app
which exposes its internal "objects" to message passing from other
programs.

Which raises the point that "all the programmers who will ever neet to
use"
a CROSS-PLATFORM "scripting language" will NOT have learned any
incarnation
of vb for that purpose.

--
-----------------------------------------------------------------------
public class JoelNeely extends FedEx {                      //       (

  String  workPhone        = "901-375-6586";                //       (
  boolean speaksForCompany = false;                  }      //     C[_]



Tue, 21 Mar 2000 03:00:00 GMT  
 Python in Dr. Dobb's Journal (Nov)

: > P.S. VB is a scripting language??

: Yep.  In its incarnation as "VBA" ("Visual Basic for Applications"), it
: is the built-in scripting language for such products as msExcel and
: msWord.  In its incarnation as "vbScript" it is a scripting language
: for Internet Exploiter.  It can also be used to drive any w32 app
: which exposes its internal "objects" to message passing from other
: programs.

Thanks for the clarification. I haven't programmed on an MS Windows 3.x
platform since writing an ODBC driver for an Ingres database a few years
back; it was an eye-opening experience working with a broken-as-designed OS.
I understand that W95 isn't nearly as broken-as-designed, but old phobias
die hard. Anyway, I just don't pay much attention to MS Windows development
tools as a result.



Tue, 21 Mar 2000 03:00:00 GMT  
 Python in Dr. Dobb's Journal (Nov)

Quote:

> Thanks for the clarification. I haven't programmed on an MS Windows 3.x
> platform since writing an ODBC driver for an Ingres database a few years
> back; it was an eye-opening experience working with a broken-as-designed OS.
> I understand that W95 isn't nearly as broken-as-designed, but old phobias
> die hard. Anyway, I just don't pay much attention to MS Windows development
> tools as a result.

Windows 95 is still broken in may ways, but its support for scripting
languages is impressive. It is surprisingly easy to talk to programs
written in other languages (usually C++) through COM. On Unix I spend
most of my time working with string streams and regular expressions. On
Windows I can just ask for full-featured objects from applications in
either the same address space or another, without me writing any C code
at all. Most of Microsoft's OS-level tools and applications have
callable interfaces and so do many other applications. Unix has ILU
which is similar, but application support is very spotty.

 Paul Prescod



Tue, 21 Mar 2000 03:00:00 GMT  
 Python in Dr. Dobb's Journal (Nov)

Quote:
> "All the programmers who will ever need to use a scripting language have
> already been born and have already learned Perl, Tcl, and VB. The
> gain in productivity in choosing Python is not large enough to offset the
> cost in learning it."

Precisely. Nonsense.

I think there is a large market out there for Python
among people who don't even consider themselves to be programmers
(witness the popularity of Numeric among chemists, physicists, etc...)
and there are even more of those.  Python is very close to ideal, IMHO,
as an easy power tool for automating computing tasks, and you don't
even have to carry around a huge reference manual everywhere you go...
People who aren't "professional programmers" will like that... Sadly,
*some*
professional programmers seem to like quirkiness and obfuscation
because it's an easy way that they can seem to be wizards without
actually knowing anything that taxes the brain.  Sorry, but this is
my analysis.   -- Aaron Watters
===
http://starship.skyport.net/crew/aaron_watters/bplustree/term.cgi?417831



Fri, 24 Mar 2000 03:00:00 GMT  
 Python in Dr. Dobb's Journal (Nov)

Quote:

> P.S. VB is a scripting language??

I'm still uncertain VB counts as a language at all :)

--
"Green Tony squeeled and I'm off to Galaxy X"




Fri, 24 Mar 2000 03:00:00 GMT  
 Python in Dr. Dobb's Journal (Nov)

Quote:


> [DDJ review of Python and "Programming Python"...]
> : [...] likes both the book and Python, but is a
> : little less optimistic about Python's future prospects than some of us.

> The argument the reviewer presents is overly brief. I will take the
> liberty of expanding it to show its weakness:

> "All the programmers who will ever need to use a scripting language have
> already been born and have already learned Perl, Tcl, and VB. The
> gain in productivity in choosing Python is not large enough to offset the
> cost in learning it."

I don't think that was the reviewer's argument.  I think a better statement
would be:

"The gain in productivity in choosing Python is not large enough to offset
the cost in learning it, given the large head start of the other
scripting languages."

The argument is more that a large base tends to continue to grow,
because new people learn what is used where they work/study...

Now, I think Python has several advantages, like NumPy, clean design, and
COM integration.  The question is will those advantages be enough to bring it
intomainstream use as much as, say, Perl.  I think the
answer is yes--look at the past 18 months and extrapolate...

kb
--

A pun a day keeps the doctor away...and everyone else, too.
http://students.cs.byu.edu/~butler/homepage.html  (updated 10/18/96)



Tue, 28 Mar 2000 03:00:00 GMT  
 
 [ 9 post ] 

 Relevant Pages 

1. Dr. Dobb's Journal Forth Article

2. FFT asm code in Dr. Dobb's Journal

3. Fortran 2003 article in Dr. Dobb's Journal

4. Source code from Dr. Dobb's Journal

5. Dr Dobb's Journal, May 1996

 

 
Powered by phpBB® Forum Software