----- Original Message -----
From: "John Porter" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Friday, June 16, 2000 9:27 AM
Subject: Re: Need some Help with Advocacy re: Java vs. Perl
> Perl needs to be thought of as a language for doing MS Windows
development,
> including the gui parts.
> (The other possibility is as a browser applet development language -- but
> that is, by all indications, NOT going to happen...)
>
> Unless and until that paradigmatic shift occurs, Perl won't be considered
> a mainstream language. It pains me to have to say it, but it's true.
>
> Perhaps Gtk is the critical link here -- for better and for worse.
The Gtk port to Win32, from the author's website, doesn't look too stable or
complete. Maybe it is, but he doesn't make it sound like he's interested in
doing more than experimentation with it. PerlGtk I didn't even know worked
on Win32. PerlQt same thing, does it work on Win32? I have the pro Qt port
and would love to see it used here, but I do have my own perl tasks to
perform and couldn't spend time porting the PerlQt module(s). I'm not aware
that it's ported already. (Is it??) PerlTk is my GUI of choice, but only by
default, and making apps in PerlTk isn't particularly interesting because I
don't know how to thread it, or even if it can be threaded, otherwise, it
would be an excellent thingy. (When it's thinking, you don't touch it.) Let
me know if I'm mistaken on any of these counts and I for one will be
developing these apps and applets out the wazoo.
One thing I would seriously like to do is redevelop my IDE (not a plug)
using Perl, especially PerlTk with threading. At present, we're forced to
consider C++ and wx as basically the only realistic alternative, Gtk being
extremely klunky-looking (and most apps' dumping enless messages to stdout
doesn't increase its attractiveness) and Qt being politically incorrect.
Java is out because Java is Java and apps don't get used if they're written
in Java simply because people avoid Java apps. Bytecode would be important
here since one part of it is commercial, but applications that would go with
it would be open.
This is one major thing that I've wanted to try - create a complete suite of
actual apps for Win32 using Perl and a GUI toolkit. It could show the Win32
world that Perl is more than CGI. I would be starting out with ICQ, IRC,
Mail, and other network things, NT admin tools, etc.
> --
>
> Another factor, IMO, is that no one sells perl modules. What little
This is _not_ a bad thing. The world needs to get away from M$'s
commercialization. The programming world was at one time almost entirely
free, until Gates screwed things up.
> perl is being sold is in the form of canned apps. Flipping through
> the trade glossies, I see lots of adds for Java beans, C++ libs,
> COM components, data blades, etc. These create the background
99% of those things are complete trash, adding to the problem of the
propagation of the pseudomyth of "You get what you pay for". Although this
statement is true for the shareware clunk that teenager with a vb for
dummies book made and markets for $15, it isn't true for the best of the
programming industry, including languages like Perl, Python, Java, C, C++,
and basically everything else.
The problem here isn't "You get what you pay for", but "any programmer
deserves to be paid for his work". He does, but in our world, payment is in
satisfaction, and our catchphrase is "The best things in life are free."
This would be almost universally true if it weren't for the marketing of
these junk toys, that really started with compuserve's shareware archive and
vbx controls.
"The best things in life are free, unless they're made for Win32 (or at
least using a M$ development tool) in which case you get what you pay for."
Not very catchy but it's the closest to true.
> radiation in the PHB's viewport: "Java is real because I see a lot
> of people selling java-based solutions!" And the ads are there
> because someone has to sell product. This puts free software at
> a huge disadvantage.
Ads for junk commercialization are still ads, and are visibility. This helps
add to myths all around, yes.
> Now here's a trick. Many (all?) trade zines include product press
> releases for next to nothing. The Perl marketing engine should
> devote itself to (among other things) writing press releases
> -- even if very little of what they say is actually new.
Convince them of that. I've paid over $600 to put ads for free perl products
and programs in TPJ, and that worked. But I'm not going to pay $1000 to put
the same ads in major rags when I get nothing in return.
> Another example: feature articles. Dr. Dobb's Journal this month
> has Yet Another Article on Rebol, which happens to say very little
> new about the language. It's essentially a sales piece.
Rebol is made by a company, and has internal integrity. Java is controlled
by sun, and has internal integrity. Python is controlled by Guido, and has
internal integrity. Perl is contolled by nobody (though Larry has the final
say), and there is no internal integrity. The p5p refuse to act as a
governing body, and nothing resembling centralization exists within the Perl
community. This has to exist before things like this can happen.
I agree that in order to increase our visiblity, apps should be written, and
am willing to do so, and have intended to do so. This is necessary to do on
Win32, since this is where the bulk of the trouble lies, but we're fighting
with fists against a brick wall of "you get what you pay for" myth which is
true mainly for the people (or community, or persons of the thought pattern)
who propagate it.
It's almost like that "fallen from grace" phrase used in christian circles.
It's used once in the Bible, specifically to condemn precicely the people
who are using it today. (Galatians...)
