WYSIWYG GUI maker V - Guido

[Warren 'The Howdy Man' Ockrassa]

[This has been xposted to the howdy-l listserv because I believe it is
very pertinent to our discussion; h-l users note this is a reply to a
poster on usenet's comp.lang.python; it is BCCed to h-l because of the
usual listserv reply-to bounce issues -- WthmO]

David Oppenheimer wrote:

> Dear Group,

My reply is usually "Howdy", so howdy!

> Looking for input here.  What kinds of features and functionalities
> would you want to have in a WYSIWYG GUI maker?

This is a wonderful question for me to respond to, since I am looking
*right this very moment* at that same question.

Brief background. I am a longtime Director hacker (Director is the
proprietary product made by Macromedia, a GUI VHLL IDE that started out
as the original presentation engine for business graphics -- it was
doing in about 1993 what PowerPoint figured out last year -- which then
turned briefly into a wonderful general purpose engine, but which lately
has become very logocentric and very commercial, to the point that
formerly-working user files are compromised by their restrictions) who
recently became very disgusted with the cynicism evinced by the Powers
That Be at Macromedia, particulaly when they effectively compromised the
insular quality of web pages which used their plugin, Shockwave, by
essentially forcing their logo and commercial for-profit site upon all
Shockwave users.

I was a very loud protestor on about a half-dozen Director-related
listservs and to some extent led the opposition there to the changes in
Shockwave... to the point that Macromedia actually backed down and
withdrew the most onerous modifications they had made. In the process I
got pretty disenchanted with the product, realizing that, in effect, the
marketing wackers at Macromedia were not interested in serving their
poweruser base, but were more interested in increasing perceived market
share (it's an old story, one we've seen before, one which effectively
kills a program and reduces it to a least common denominator influenced,
I suspect, mostly by Redmond).

In addition to getting tired of fighting so hard to keep Director a more
or less pure IDE, I became distrustful of the decisions made by the
parent company, to the point that I believe similar bad mistakes will be
made in future, mistakes which will effectively corrupt the product to
the point that it is used solely for entertainment content. I don't
think I am the only one to feel this way.

Those of you who know a little of Director might be surprised to know
that, for a couple years anyway, it really was a hacker's IDE, that it
didn't actually have to be a whiz-bang toy, which is what, apparently,
the new blood at Macromedia has in mind for their product. Argh!

As an example of the more recent for-profit bent of the company,
Macromedia recently announced the release of an Xtra developers' bundle
which presumably makes net DL of Xtras easier, but which costs US$50 to
acquire. Problem is that without the third-party Xtras made by C++
hackers, Director would have almost none of the extensibility it has (it
relies on Xtras for almost EVERYTHING now, including basic net ops and
file read/write to/from disk).

So in order to create extensibility packages for Director, which help to
make Director more marketable for its parent company, you have to own
and be proficient with Director, have to be a decent C++ hacker, *and*
pay Macromedia money for the "privilege" of extending their tool for
them. Sound familiar? Can anyone say Bill?

(The VP in charge of distribution recently told me, in essence, that if
Macromedia had wanted to ask their powerusers what should be done with
the product, they would have. Meaning of course that they didn't even
consider asking the powerusers... worse; actually they considered asking
the powerusers and discarded the idea in favor or market-driven eye
candy. Whence the eye candy, though, ultimately? Hackers, of course! Who
pushes the envelope harder?)

This recent turn of events was so hateful to me in its intent that I
decided it was time to write an open source VHLL IDE like Director, one
which could *never* be owned by *any* logo-bearing company.

Having dilettanted about with various VHLLs before that time I knew of
Python and have turned to it exclusively as the focus of my time and not
inconsiderable energies. (Anyone who knows me online knows I tend to be
rather consistent in the way I pursue an idea, in much the way that a
pit bull is consistent in how it handles an attacker. ;)

I am in the process right now, on my own litle listserv, of cobbling
together something like Director's IDE but with much more poweruser
access, in Python. I have no doubt that Python is up to the task. As I
learn more about it I find new things which can be used to create the
very thing you seem to be asking after. (And I am continually amazed at
the richness in the Python environment; it outperforms Director's
internal language, Lingo, by at least one order of magnitude; things
near impossible in Director are ludicrously easy in Python.)

You and *all* interested parties are invited to hit nightwares (my site)
and sub to howdy-l, where discussion of this very thing is taking place
daily now. Experts in Python would be most welcome, of course, since
such folk would be best in a position to indicate, at least, directions
our explorations can take. Of course *anyone* with an opinion is
desired; the list sub is about 35 now but can definitely expand. I need
input! We need input! (Lurkers are welcome too.)

While on nightwares, please do sample some of the other sections, which
serve as tutorials on Director, show off some of its capabilities, and
which include links to places you can go to get more info on Director.

The project I have initiated, working title kurosawa, is meant to be an
opensource GPL answer to Macromedia's stranglehold on the rich media
development marketplace, and I can't see any reason at all why Python
isn't up to the task; it already has waaaaaaaaay more power than
Director, which costs US$700 per platform and doesn't even run on *NIX.

(I'm doing kurosawa because I believe *everyone* should have that kind
of power and I do not want it to be a Director clone -- I want it to be
much, much better.)

I can go into terabytes worth of more depth on this at request, but
since my ISP has limited me to 500 Mb for my domain quota, and since I
have about 70 Mb in use now, I'd have to slice up the details in 300 Mb
increments, which is of course why I am keeping my reply so terse here.
:)

I know I'm new here on c.l.p, but I am not new to VHLL IDEs, and I think
what you are asking about meshes perfectly with what I'm working on
right now, with an elite group of very intelligent and very brave
volunteers.

I also know I am coming at this from the other side -- I'm not terribly
good with "real" compilers like C; I prefer the very high level access I
get from IDEs such as Director, AuthorWare, MetaCard, and of course
Python. (I do not have a pathological terror of curly braces or explicit
variable declarations but, given the power of interpreters, I wonder why
anyone really thinks they are necessary any more beyond a certain level
of neo-Luddite purism.)

I also know it is possible for these two camps to merge, and that by
golly is what I have in mind with kurosawa.

Thanks for asking. :D

PS: You can see my posting history on Direct-L, a major Director
listserv, by surfing the Maricopa archives I have linked off nightwares
under others --> maricopa, especially in the last 6 months, to get a
flavor for the astounding and profanity-laden levels of frustration I
grew with what I saw as an increasingly-monolithic dictatorial engine.
If you're interested in checking me out, that's one place to get a real
slice.

Best to all!

--
    warren ockrassa | nightwares | mailto:warren at nightwares.com
         --    n   i   g   h   t   w   a   r   e   s    --
      director faq    lingo    tutorial    free files    links
                    http://www.nightwares.com/