Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Groups keyboard shortcuts have been updated
Dismiss
See shortcuts

The wonders of Dragonfly

27 views
Skip to first unread message

Mr. Anderson

unread,
Jan 17, 2014, 7:24:56 PM1/17/14
to
Mr. Doe,

I recently came upon this:
http://b09s25le.corenetworks.net/node/2724

Very nice of Mr. Butcher to share some code with you. I wonder why he
doesn't speak much about Dragonfly in public these days.

Dragonfly seems just the thing to help get me coding by voice, if only
I could even begin to understand how to use it. Your example code was
helpful. I wonder, are there any other Dragonfly snippets you have
laying around that might be interesting?

Most recently I attempted to get a working sample going which employs
his base rule class, named "Rule". I have failed to do so. Here is my
first feeble attempt:
class RuleRule(Rule):
def __init__(self, exported=True): # Make it a top-level command.
pass

def _process_recognition(self, node):
print "Rule recognized."

I'm wanting to get working examples of some his classes to show how
they could be used, so that I may even eventually do something
worthwhile with them. I've seen a few five videos where Dragonfly is
used, only two of them being of much use to me. The rest of them were
either a bit lame or were just showing off without teaching anything.
That's what's frustrating about Tavis Rudd for me. He shows off but
doesn't even give an inkling as to how he accomplishes his goal.

Regards,
Mr. Anderson

John Doe

unread,
Jan 17, 2014, 11:01:40 PM1/17/14
to
Mr. Anderson <matri...@yahoo.com> wrote:

> Mr. Doe,
>
> I recently came upon this:
> http://b09s25le.corenetworks.net/node/2724
>
> Very nice of Mr. Butcher to share some code with you. I wonder
> why he doesn't speak much about Dragonfly in public these days.
>
> Dragonfly seems just the thing to help get me coding by voice,
> if only I could even begin to understand how to use it. Your
> example code was helpful. I wonder, are there any other
> Dragonfly snippets you have laying around that might be
> interesting?

What sort of snippets?

Before using Vocola... I put everything (all of my Dragonfly macros)
in one file. Putting the global macros at the bottom of the file
causes program specific macros to take priority.

I don't do "grammar" or whatever you guys call it. I just record
macros and stick them in a file. Trying to make the voice activator
as short as possible... one, two, or three syllables.

> Most recently I attempted to get a working sample going which
> employs his base rule class, named "Rule". I have failed to do
> so. Here is my first feeble attempt:
> class RuleRule(Rule):
>
> def __init__(self, exported=True): # Make it a top-level
> command.
> pass
>
> def _process_recognition(self, node):
> print "Rule recognized."
>
> I'm wanting to get working examples of some his classes to show
> how they could be used, so that I may even eventually do
> something worthwhile with them. I've seen a few five videos
> where Dragonfly is used, only two of them being of much use to
> me. The rest of them were either a bit lame or were just showing
> off without teaching anything. That's what's frustrating about
> Tavis Rudd for me. He shows off but doesn't even give an inkling
> as to how he accomplishes his goal.

You said you got it to work already. What happened?

I don't like the idea of going with Dragonfly, because the original
author just isn't here anymore. At the same time, the Vocola guys
are around. Mainly Mark Lillibridge, but Rick Mohr might be lurking.
And the Natlink "administrator" guy is currently active. So is the
guy that wrote the main part of Natlink (but I'm not sure where he
comes down, either Vocola or Dragonfly, could find out). Speaking of
people who have shared code with me... Mark Lillibridge has been
exceptional in that respect, one Vocola 2 solution after another at
(or at least after) my request.

I agree about Tavis Rudd. Has he done anything with it since that
presentation? Some people fall out of love with speech recognition
after the fascination wears off. It never ceases to amaze me that
there aren't a bunch of real programmers interested in this speech
activated scripting stuff. But oh well.











One bit of good news, if I'm not mistaken, the Natlink administrator
might make installing Dragonfly easier (something to do with
removing his pet project from the installer), even if the original
author never gets around to it.

Mr. Anderson

unread,
Jan 18, 2014, 1:20:27 PM1/18/14
to
>What sort of snippets?
Anything at all really. Preferably stuff that demonstrates more of the
classes in Dragonfly. The documentation simply does not cover much
ground regarding actual use.
>Before using Vocola... I put everything (all of my Dragonfly macros)
>in one file. Putting the global macros at the bottom of the file
>causes program specific macros to take priority.
If you'd like to share this file, that'd be great, but I won't nag you
about it if you don't.

>You said you got it to work already. What happened?
I'm talking specifically about the Rule class. Since you don't
approach it from a programming perspective, you probably don't even
care about that.

I suspect most people would say to just use CompoundRule or
MappingRule, which apparently derive from Rule. However, it would be
nice to have a greater understanding and be able to use the base Rule
class directly in an example command.

>
>I don't like the idea of going with Dragonfly, because the original
>author just isn't here anymore.
Yeah, the last I saw him post was in April of 2013. I'd like to see
him promote his framework more, because I think it has capabilities
that can only otherwise be accomplished with NatLink directly.

>the Vocola guys
>are around. Mainly Mark Lillibridge, but Rick Mohr might be lurking.
>And the Natlink "administrator" guy is currently active. So is the
>guy that wrote the main part of Natlink (but I'm not sure where he
>comes down, either Vocola or Dragonfly, could find out). Speaking of
>people who have shared code with me... Mark Lillibridge has been
>exceptional in that respect, one Vocola 2 solution after another at
>(or at least after) my request.
Yeah, Mark seems like a very nice guy.

>I agree about Tavis Rudd. Has he done anything with it since that
>presentation? Some people fall out of love with speech recognition
>after the fascination wears off. It never ceases to amaze me that
>there aren't a bunch of real programmers interested in this speech
>activated scripting stuff. But oh well.
I agree, and I think a guy like Rudd could help a lot if he would just
release some example code. But oh well.

>One bit of good news, if I'm not mistaken, the Natlink administrator
>might make installing Dragonfly easier (something to do with
>removing his pet project from the installer), even if the original
>author never gets around to it.
Yeah, I'm afraid to upgrade to the new version that fixes the CCR bug
because of what Quintijn said about not being sure how it affects
Dragonfly right now.

John Doe

unread,
Jan 18, 2014, 2:08:27 PM1/18/14
to
http://sdrv.ms/1cqAvyt

Let me know whether it works. If not, there is probably some simple
fix. You put those in the Macrosystem folder. But that was before the
folder structure was changed, I think. You might have to change some
paths, I dunno.

How is voice-activated scripting useful to me? As you can see by the
global macros/scripts, I use it systemwide a lot for copying,
cutting, pasting, and other keystroke shortcut repetitive Windows
tasks. Much of my stuff is done by positioning the pointer and then
speaking the word. Like "point and script" instead of "point-and-
click". Very useful all over Windows. Depending on the game, it can
be extremely useful for gaming. I couldn't play half as well without
it. It takes most of the clickfest out of the game.

Besides all of the other planned stuff, something that's potentially
useful is enhancing the ability to make and execute temporary
scripts.

As Christo pointed out, voice lessons can be useful if you do this
stuff heavily. I believe there is a risk of repetitive voice injury,
similar to repetitive strain injury, because you are repeating the
same words umpteen times, it's not like normal human speech.

It also helps if you use speech recognition for typing/dictation.
That is a great benefit by itself, if you write a lot.

John Doe

unread,
Jan 18, 2014, 2:13:42 PM1/18/14
to
Recently discovered a useful tactic for working with text.
Oftentimes, the insertion point can be before the word under the
pointer instead of precisely under the pointer. So the macro begins
with a (click) plus (control left) instead of just a (click). That's
very useful because it greatly eases the pointing requirement. I've
implemented it on several macros, but I think it can be used with
more. I don't believe that tiny adjustment was implemented in the
Dragonfly scripts on my Sky Drive, though.

http://sdrv.ms/1cqAvyt

Mr. Anderson

unread,
Jan 18, 2014, 11:52:30 PM1/18/14
to
Thanks for sharing.

Mr. Anderson

unread,
Jan 19, 2014, 1:58:41 PM1/19/14
to
You could be right that Vocola has everything to offer that Dragonfly
has, but then why would Rudd be using Dragonfly is my question.

Maybe someone will chime in on the mailing list.

Thanks

John Doe

unread,
Jan 19, 2014, 8:49:08 PM1/19/14
to
Mr. Anderson <matrixlove yahoo.com> wrote:

> You could be right that Vocola has everything to offer that
> Dragonfly has, but then why would Rudd be using Dragonfly is my
> question.

You can ask that person...
0 new messages