MS-Agent 3.0
Stephanie Heck
I am thinking that MS Agent software would be much more popular if it were
capable of supporting audio formats in addition to .wav The current
capacity to support wav files is great, but a file format that has tighter
compression would make MS Agent presentations more palatable to the average
end user.
Microsoft owns copyrights on several Codecs and audio compression
technologies. Perhaps this could be incorporated into the next version of MS
Agent.
Stephanie Heck
webm...@sheck.com
http://www.sheck.com/about.htm MS Agent Page
James Driscoll Jr
"Stephanie Heck" <sh...@softhome.net> wrote in message news:OQyQ8ARq$GA.187@cppssbbsa04...
Remy Lebeau
MSAgent 3 will not be bitmap-based anymore, or so I've heard.
Gambit
James Driscoll Jr <spac...@resistanceisfutile.com> wrote in message
news:OmPiheTq$GA.256@cppssbbsa03...
Rob Lindman
> I am thinking that MS Agent software would be much more popular if it were
> capable of supporting audio formats in addition to .wav
It is possible to get the current version of MS Agent to 'speak' audio
formats other than those specifically supported, but not by conventional
means. Note also that there are a few WAV codecs which don't function with
Agent.
MS Agent technology would be more popular if there were fewer complications
to implementing MS Agent, and better software that implemented MS Agent.
"Remy Lebeau" <gamb...@gte.net> wrote
> ...the new character file format for MSAgent 3 will not be bitmap-based
anymore, or so I've heard.
I'm hoping Microsoft does not abandon their current methodologies for Agent
by ditching bitmap support in the format. There are some very cool
possibilities with the current format which to my knowledge are not being
explored by most developers.
Also, there is now a fairly substantial library of characters in the current
format on the web, to render (no pun intended) these incompatible would
probably not be favorable with the developers of these characters unless
there is a straightforward upgrade path. Keep in mind that Warner Bros spent
$50,000 to get Bugs and Daffy made. While they might have another $50,000 to
spend, they won't necessarily be inclined to do so.
A 3D mesh / skin scenario would be a good addition to the capabilities of
Agent, but not at the expense of eliminating the already extremely useful
format.
Rob
Lightbee
more it would be to much and any less would just make for a bad image. the
bmps I hope will be supported in the new version of agent if they use
some other image format.? and you can always make an exe that sets it's
self up and has other format audio in that would go to the viewers media
folder but then that is taking the chance that the viewer will not erase
the audio for your agent. I think if they would allow time for making
software that is free and is a lot easier to make the agent images with and
let the motion software alone (except for any bug fixes that might be
needed) the agents would be more wide spread across the net. for we all
get the same acs files but unless you know where the acf files are the
viewer has to leave your site and hunt the acs file in order to see the
agent if they do not have it.
and that is a major pain in the.....
on another note. a patch to enlarge the size limit of the font folder
would be very much welcome. mine keeps filling up and crashing the system.
roving cowboy
"Rob Lindman" <thegre...@hotmail.com> wrote in message
news:OXWckcaq$GA....@cppssbbsa02.microsoft.com...
Glenn Saunders
profile contract awarded to a custom agent creator, unless that parrot
that L&H sells counts, is that MS Agent?
When MS Agent was conceived, the lowest common denominator graphics
card on the PC and the CPU was not close to being up to the task of
doing realtime 3D characters that would look halfway decent. I
understood the compromise, but today the existing bitmap structure to
MS Agent is clearly outdated. I mean, it's not even a 24-bit format.
It's fixed-palette 8-bit, and it shows. One can indeed implement a
JPEG-based 32-bit format. Macromedia Flash does this when exporting
bitmaps originally imported as 32-bit PNGs, for instance. No fringing
(better than Agent does when rescaling, BTW). Still, if you look at
Agent as an outgrowth of game technology, it's closest cousin would be
Mortal Kombat or Donkey Kong Country, not more modern titles like Tomb
Raider. Other than its ability to overlay on top of windows, in 2000,
MS Agent looks like old tech.
Nevertheless, bitmap is too heavy for the internet. That's why Flash
is so popular, because for most things you can get the job done with
just Vector. And it helps that vector is resolution-independent.
There has yet to be a truly viable 3D vector-based real-time character
system like agent, outside of SGI workstations maybe, but the
technology has been available to do such a character for at least the
last 3 years.
Pulse3D is closest, but it's just not there.
The audio support of MS Agent is also outdated. Yes, many subcodecs
within WAV won't play properly by agent. Windows Media Audio would
be great for agent. Being able to hook agent to the Windows Media
Player for lipsync would be great.
So I am all for making radical changes to MS Agent for version 3, if
it ever comes out. A vector format would invovle wholescale changes
to the animation system. Why store scripted commands when you could
execute more specific "relative" commands in realtime to twist limbs
and so on (more like the command system of things like Alice3D). I
believe most game engines these days also employ animation blending
and localized animations so you could have characters crouch and then
punch instead of having a standpunch and a crouchpunch animation.
Then what about character<>character interactions? Collision detect?
Props? Physics? Direct user control?
Didn't they tinker with this before with Liquid Motion? That was a
dog, wasn't it?
With the apparent death of VRML (3DML does not seem to be doing much
better) if Microsoft would really like to take over this niche,
rolling MS Agent into a generalizable "windows only" 3D engine for the
web would be advisable.
Phil Jackson
Some of those apps like Alice, Shout3D, Anfy3D would wipe out MsAgent if
only TTS was in the mix. The external problem to arguments over flash
versus bitamps etc, probably comes down the intellectual copyrights of the
various technologies making it too much of an administrative nightmare to
proceed with. The other lousy thing is the dependancies of having to
download and install plugins, activeX controls, or even a specific browser
just to see the effects. Personally, I'm finding that I'm more impressed by
some of the java-applets than MsAgent simply due to the fact that the end
user doesn't have to do anything. Maybe some bright spark will eventually
develop one that takes advantage of SAPI.
- Phil Jackson.
"Glenn Saunders" <cybp...@earthlink.net> wrote in message
news:38fea6da...@news.earthlink.net...
Grandpa Jim
http://www.netsage.com/reversi/reversilogin.htm
It's been out for a year now.
"Rob Lindman" <thegre...@hotmail.com> wrote in message
news:OxdZJH$q$GA.265@cppssbbsa03...
: Glenn Saunders wrote
: >When MS Agent was conceived, the lowest common denominator
graphics
: >card on the PC and the CPU was not close to being up to the task
of
: >doing realtime 3D characters that would look halfway decent.
:
: > the existing bitmap structure to MS Agent is clearly outdated.
I mean,
: it's not even a 24-bit format.
: >It's fixed-palette 8-bit, and it shows.
:
: I do agree that enhancing the pallette to add more colors may be
cool, but
: slow computers are still out there, as are lower grade video
cards.
:
: Don't forget that most people really don't even know how to adjust
their
: monitors resolution to a higher setting unless it's explained to
them. We in
: this group tend to be much more computer savvy, but you can't
overlook the
: fact that most people don't tweak with their computer because they
are still
: intimidated by the possibility that they can mess things up.
Ignore that
: fact and watch your product die.
:
: For a brief period of time I experimented with doing on location
tech
: support for end users and companies. The majority of people were
running
: 640x480x16, even when they had high end video cards. I have a 32
meg video
: card, and I was astonished when one guy had a 64 meg card and was
running in
: the lowest resolution. He didn't know it was that cool, and he
didn't know
: how to take advantage of it. This is representative of the
majority of the
: people using the Internet, not software developers, but normal
humans who
: couldn't care less about palette depth. These people will never
refer to
: Merlin as an "ms agent", but will just think of him as Merlin, the
little
: wizard in the computer. These are the people who will buy the
products we
: design, and they are the ones Microsoft sells their OS to.
:
: Agent is here to help people who don't understand technology
become better
: able to accomplish tasks with their computer, not just so we can
geek out.
:
: > No fringing (better than Agent does when rescaling, BTW).
:
: I'd love to see improvements in the fringing area, too, but the
most
: problems I have with this are because I don't have a decent
palette
: optimizer. I looked around for a long time to find a free one, and
in the
: end I had to piece together my own solutions, which isn't all that
great. It
: would be nice if the scaling worked without creating glitches in
the window
: region. I'm hoping this is improved in Win2k and WinME because of
the newer
: DirectX layer.
:
: >There has yet to be a truly viable 3D vector-based real-time
character
: >system like agent, outside of SGI workstations maybe, but the
: >technology has been available to do such a character for at least
the
: >last 3 years.
:
:
: I'm not certain Agent is intended as a 'gaming technology' or a
'graphics
: technology'. I believe it's intent was in the area of user
interface
: enhancement. It suits this fine. It would be a pain to deal with
all of the
: 3D vector based nonsense just to make a little guy.
:
: >The audio support of MS Agent is also outdated. Yes, many
subcodecs
: >within WAV won't play properly by agent.
: > Being able to hook agent to the Windows Media Player for lipsync
would be
: great.
:
: My research indicates that the majority of these incompatible
subcodecs were
: created after the current release of Agent. Keep in mind there are
numerous
: ways you can do this with and without Agent. Microsoft provides
huge numbers
: of controls and technical samples which actually demonstrate steps
necessary
: to do things of this sort. If you are familiar at all with audio
: programming, you can design your characters to be compatible now.
There are
: other ways to fake this sort of thing too.
:
: I would prefer asynchronous output support for TTS/Audio Files and
the
: ability to specify stereo options, but these may be TTS only
functions and
: it could be L&H's fault. Also, make the damn lips move, even if
they guy
: can't talk, and use some sort of a timer. It kills me that the guy
talks but
: his lips don't move when I'm playing other audio.
:
: >So I am all for making radical changes to MS Agent for version 3
:
: I am entirely against 'radical changes', and my belief is that
Microsoft
: generally doesn't do this sort of thing, especially with
technology that's
: in somewhat wide use on the net already. Agent has so much
flexibility it is
: nearly a universal improvement to Microsoft's operating system.
:
: If they really want to kill any and all momentum to use this
technology,
: they'll listen to you, but I don't see any real benefit to what
your
: proposing. If they want to improve gaming technology, they should
just start
: another group, and they probably already have one.
:
: > I believe most game engines these days also employ animation
blending
: >and localized animations so you could have characters crouch and
then
: >punch instead of having a standpunch and a crouchpunch animation.
:
:
: Again, I don't think Agent was designed with games in mind.
Enhancing the
: animation capabilities would be cool, but not inherently necessary
to
: Agent's current implementation.
:
: There are much better technologies for creating games out there.
If you want
: to create games, you should investigate them. If you want to add
speech
: capabilities to your game you can use the other controls provided
by
: Microsoft in their speech SDK. These will allow you to use similar
effects
: to Agent without waiting for v3 or forcing MS to destroy the
benefits of
: Agent's extremely usable format.
:
: Agent is good because it has a small download footprint. Making it
into more
: than it is would sacrifice a lot of it's benefit. My 'surveys'
indicate most
: people dislike downloading at it's current size, why make it
bigger?
:
: >Then what about character<>character interactions? Collision
detect?
: >Props? Physics? Direct user control?
:
: All of these things can be accomplished with Agent in the current
version.
: If you need a prop, make a character out of it. Collision
detection is
: simple, Agent's characters are essentially windows with
rectangles. You can
: check the areas of the rectangles against the other rectangles.
Remy said
: something similar to this with the DOM and comparing locations to
document
: objects.
:
: Not sure what you mean by physics exactly, but you can manipulate
the Top
: and Left properties in any fashion that is necessary to create the
: appearance of reaction to outside forces. If you want a character
to bounce,
: fall, lift off, tween, or anything else, you can do this if you
are a
: sophisticated developer.
:
: AFAIK, Agent isn't designed to be an all in one graphics solution
to replace
: Flash or a vector system. The benefit to Agent and it's use on the
web is
: that it can be used to supplement these other technologies. Making
it top
: heavy would be a mistake in my opinion.
:
: Rob
:
:
:
:
Grandpa Jim
I am on your side.I like MS Agent like it is.
"Rob Lindman" <thegre...@hotmail.com> wrote in message
news:uz9Qft$q$GA.213@cppssbbsa03...
: Grandpa Jim wrote in message ...
: >Try this MS Agent game.
: >http://www.netsage.com/reversi/reversilogin.htm
: >It's been out for a year now.
:
: Yes, Agent CAN be used in games, but it doesn't mean that it was
: specifically designed to do so. I've been around long enough to
know about
: the game you are pitching here.
:
: Additionally, the point of my argument isn't that Agent shouldn't
be used in
: games, it's that current functionality and usability and size
benefits
: shouldn't be discarded just for the purpose of games. I don't get
the
: feeling you considered my entire post before your reply. If
anything you've
: strengthened my position that it's not necessary to totally modify
Agent's
: format to make it useful in game situations by demonstrating that
there are
: games which use the technology.
:
: Game development has advanced sufficiently in other areas. Some
people just
: think that one technology should be the answer to all of their
software
: desires and in my opinion that's ludicrous. An indication that I'm
correct,
: the game you've mentioned here would be outdated by a total
modification of
: Agent's format. Would you be happy if Microsoft suddenly made this
game
: non-functional?
:
: Agent currently retains backwards compatability with previous
versions of
: Agent, at least v1.5. I'm inclined to believe that Microsoft
generally tries
: to keep their products somewhat compatible with previous versions,
that's
: why I don't think we'll see a 'totally different' agent, unless
it's so far
: out there that it's impossible to deny the benefits.
:
: Far be it from me to tell people what they can and can't do with
Agent, I
: just don't want to see Agent's format modified so extensively that
currently
: unrealized benefits are destroyed.
:
: I will be filling people in on what I mean shortly, I am in the
process of
: setting up my domain to demonstrate this and offer some of my
innovations
: regarding this technology for everyone to see.
:
: Rob
:
:
Rob Lindman
Rob Lindman
>http://www.netsage.com/reversi/reversilogin.htm
>It's been out for a year now.
Yes, Agent CAN be used in games, but it doesn't mean that it was
Glenn Saunders
<thegre...@hotmail.com> wrote:
>Don't forget that most people really don't even know how to adjust their
>monitors resolution to a higher setting unless it's explained to them. We in
>this group tend to be much more computer savvy, but you can't overlook the
>fact that most people don't tweak with their computer because they are still
>intimidated by the possibility that they can mess things up. Ignore that
>fact and watch your product die.
Plugins should never be targetted to barely computer literate end
users. They are an extra feature for those who know how to properly
enable them.
>how to take advantage of it. This is representative of the majority of the
Store bought machines are preconfigured to 800x600 at 24-bit color,
typically.
>Agent is here to help people who don't understand technology become better
>able to accomplish tasks with their computer, not just so we can geek out.
Then maybe Agent application developers should contain readmes on how
to right click on the damn desktop to change resolution and screen
depth. Problem solved.
>I'd love to see improvements in the fringing area, too, but the most
>problems I have with this are because I don't have a decent palette
Fringing is introduced by Agent even if the frames are immaculate.
>region. I'm hoping this is improved in Win2k and WinME because of the newer
>DirectX layer.
I doubt it. It has nothing to do with the OS. It has to do with the
AgentServer control's rendering process.
>I'm not certain Agent is intended as a 'gaming technology' or a 'graphics
>technology'. I believe it's intent was in the area of user interface
>enhancement. It suits this fine. It would be a pain to deal with all of the
>3D vector based nonsense just to make a little guy.
"Intelligent Agents" themselves don't even need a visibible UI.
Technically speaking MS Agent isn't even a true agent, it's just the
front-end of one. I think Microsoft focused on the wrong end of the
technology, the look and feel, as opposed to the brain aspect. But if
they are going to focus on the look and feel, they could have gone
further by now.
>My research indicates that the majority of these incompatible subcodecs were
>created after the current release of Agent.
Not after the release of Agent 2.0. Microsoft had more than enough
time to update the control. Agent 2.0 was not a very progressive
update. Besides, there is enough pre-existing component-based OS
support for playing various media types in Windows that Agent could
leverage that they almost had to WORK to make it incompatible. Agent
already has dependencies on an external TTS and speech recognition
engine. There is no problem with it being dependent on things like
Windows Media Player or other components the user is likely to have.
>to do things of this sort. If you are familiar at all with audio
>programming, you can design your characters to be compatible now. There are
>other ways to fake this sort of thing too.
If you could, you'd wind up rewriting part of the core of what Agent
is designed to do internally.
>I am entirely against 'radical changes', and my belief is that Microsoft
>generally doesn't do this sort of thing, especially with technology that's
Microsoft sometimes makes radical changes. Windows 3.1 -> 95, 98 ->
2000. DirectX has seen very rapid development and enhancement.
Microsoft has rapidly adopted support for new port standards like USB.
>in somewhat wide use on the net already. Agent has so much flexibility it is
>nearly a universal improvement to Microsoft's operating system.
I don't see Agent as an improvement to the operating system at all,
since Microsoft itself only uses it as an online help system.
>If they really want to kill any and all momentum to use this technology,
>they'll listen to you,
I happen to feel that some of MS Agent's limitations are the reason
why it hasn't been used more.
To be afraid to enhance Agent is to be afraid of progress... And
that's STUPID in the computer industry.
Many technologies RELATED to MS Agent are in heavy development.
Graphics engines, speech recognition, and so on. Yet MS Agent stands
largely static watching these technologies progress, rather than
taking advantage of them. This is rather sad, I think.
MS Agent 2.0 doesn't have to die, keep using it if you like, but I'm
more than ready for the next generation.
>Again, I don't think Agent was designed with games in mind. Enhancing the
>animation capabilities would be cool, but not inherently necessary to
>Agent's current implementation.
If Agent is about having characters emote through action, then any
technology which helps them emote better is important. Gaming
technology offers this.
Let's have some perspective here...
Even the original Playstation Toshinden's characters emote better in
their victory gestures than Microsoft agent, and that game is five
years old and ran on a 33mhz RISC processor which can not be fully
emulated by today's PCs.
>to Agent without waiting for v3 or forcing MS to destroy the benefits of
>Agent's extremely usable format.
Dude, people still open DOS windows. Just because something better
comes along doesn't mean the old stuff disappears. It just slowly
fades away.
>Agent is good because it has a small download footprint.
It has yet to be determined that these new features would bloat its
footprint too much. Besides, broadband is on the way.
>All of these things can be accomplished with Agent in the current version.
>If you need a prop, make a character out of it.
How do you get a character to pick up a prop? Throw it? Juggle it?
Have it appear in front of or in back of the character? There is no
way to control Z-index. How do you have a character pass behind an
HTML element? Lots of limitations. Agents instances are islands of
graphics and any attempt to overcome that isn't worth the monumental
effort...
How do you create a 3D background and have the agents "work" within it
effectively? I don't think you can, not with a sprite-based engine.
>AFAIK, Agent isn't designed to be an all in one graphics solution to replace
However, Agent has been used beyond its original design
specifications. At I Pyxidis we were using them as virtual actors
performing mostly scripted lines in realtime, similar to how Pulse3D
uses its characters. These features would have been very useful for
that.
Glenn Saunders
<thegre...@hotmail.com> wrote:
>Yes, Agent CAN be used in games, but it doesn't mean that it was
>specifically designed to do so. I've been around long enough to know about
>the game you are pitching here.
That's the problem. Agent's conception was too narrow, and hence its
usefulness was curtailed by the fact that it _isn't_ a generalizable
display engine.
While Agent really isn't that useful in writing games (or creating
scripted virtual characters that interact with eachother and
background), GAME engines are up to the task of being applied to
creating AGENT UIs, which is all MS Agent really is.
Game engines encompass the requirements of an Agent UI and go far
beyond them! So you see, it would not be a bad thing for a next-gen
Agent control to be done this way--it would actually give it more of a
chance of being a success as it would be useful in so many arenas, not
just the strict notion of MS Agent.
So making MS Agent more into a generalizable display engine, it
retains its usfeulness as a UI, but also opens up all sorts of other
applications in one fell swoop.
Let me give you an example. Using MS Agent in multiuser chat.
You get all the agents on a web page talking. First of all, no
overlapping dialogue possible. That kinda sucks. Secondly, all the
characters are 3D rendered sprites all perspective-flat across the
entire screen. You can not impose any sort of workable 3D background
for them to be in for chat-room atmosphere. Maybe you can do a Diablo
sort of thing with Isometric, but it's still pretty lame. You can't
have the characters pick up background elements. It's a pure
backdrop, noninteractive. You can't really have characters walk from
place to place. They "glide" with movetos. Basically they just look
like they are hovering over the page, not a part of it.
The TTS and the real-time animation make Agent really ideal for
multiuser chat over other off-the-shelf plugins like straight VRML or
the cumbersome and non-TTS Pluse-3D, but the fact that each character
has no conception of their environment or what's going on with other
characters hinders all sorts of applications like these.
What if you create a MASH script with two characters talking to
eachother. First of all, unless the character has a suite of
animations in 90' angles both left and right, they will appear to be
in a play with them emoting straight ahead and not looking at
eachother. Because it's bitmap based, the "download footprint" as you
call it, would skyrocket to give the character enough positions to
allow it to properly "emote" in all directions, sit down, roll around,
and so on.
MS Agent was designed so that its footprint would be manageable as
long as the animations all branched off of a forward-neutral
perspective. Once you want a truly generalizable character that can
"emote" better than a block of wood you will be spending all your time
rendering frames to create all the animations, and the end-user will
be downloading animation all day.
Even a larger initial download for a realtime-3D agent engine would be
made up MANY TIMES OVER in the flexibility of simply ordering the 3D
character to turn 90' left or right and then firing off the other
animations from the new perspective.
There are plenty of games that border on Agent technology. For
instance, virtual pets like Pokemon, Dogz and Catz, and so on, or that
talking Fish they have been talking about on the Playstation 2 that
reacts to speech input. The gap isn't as wide as you might think it
is. It makes sense to leverage game know-how for MS Agent.
T-man
HTML+TIME has been updated twice since the IE5 went final. The XML parser has been updated twice in the last four months. The
scripting engines have been updated twice since then, too, and IE5.5 is in its second public beta. DirectX is two full versions up
on Agent 2.0, and Media Player is three. MSAgent is not only the slowest moving of any of the browser technologies, it's the
strangest ActiveX I work with. Since the brilliance of the web-release mentality doesn't seem to have permeated the darkness of the
MSAgent closet, and MS is perfectly willing to pull the plug on products without notice. I'm guessing that the DOS box may soon be
welcoming Genie home. . .
T
"Glenn Saunders" <cybp...@earthlink.net> wrote in message news:3902911b...@news.earthlink.net...
Rob Lindman
>Technically speaking MS Agent isn't even a true agent, it's just the
>front-end of one. I think Microsoft focused on the wrong end of the
>technology, the look and feel, as opposed to the brain aspect. But if
>they are going to focus on the look and feel, they could have gone
>further by now.
Microsoft creates software components for developers to build upon. This
creates a market for Microsofts developer tools. It's up to the developer to
use the tools in their own innovative way to create markets for their own
products or services. Would you prefer that Microsoft simply makes
everything and therefore defeats the purpose of designing software entirely?
There are nuances to Microsoft's strategies which are perceptable to those
who have watched the software industry for a considerable length of time. I
could explain to you why I believe they do what they do, but I'd rather not,
because it would appear you would rather they just designed everything, and
who am I to argue with you?
> Plugins should never be targetted to barely computer literate end
> users. They are an extra feature for those who know how to properly
> enable them.
Well designed software should accomodate users who don't care about their
computer aside from it's impact on their daily lives. Marketable software
targets people, not geeks. Everything else dies, because geeks would rather
outgeek each other. And geeks traditionally don't buy software.
> Store bought machines are preconfigured to 800x600 at 24-bit color,
> typically.
When Grandma resets it to 640x480 somehow, she can't get it back without
asking someone for help, or paying the computer store $60 to do it for her.
Grandma is one potential user of Agent, and probably one of the most
important.
> Then maybe Agent application developers should contain readmes on how
>to right click on the damn desktop to change resolution and screen
>depth. Problem solved.
A smart developer writes a program for Agent that says "Grandma, your
computer can be changed so that your baby grandson's picture looks better.
Would you like me to change this for you?". That's how you solve the
problem. Grandma doesn't know or care about the damn README. Neither do I. I
just want grandma to give me a dollar, so I can eat, and write more
software. Why is Grandma going to give me a dollar if I make her life
difficult?
>I happen to feel that some of MS Agent's limitations are the reason
>why it hasn't been used more.
Maybe. Perhaps, as I stated, most people are intimidated by their computer,
and most people couldn't care less. Perhaps overcoming Agent's limitations
is one way you go about designing an innovative product and create a new
market.
>How do you get a character to pick up a prop? Throw it? Juggle it?
>Have it appear in front of or in back of the character? There is no
>way to control Z-index. How do you have a character pass behind an
>HTML element? Lots of limitations. Agents instances are islands of
>graphics and any attempt to overcome that isn't worth the monumental
>effort...
There are ways to control Z-index. Props can be accomplished by intelligent
animation design. Characters aren't designed to be behind HTML elements,
they are supposed to be in the foreground of all applications, but you could
do this if you put some thought into it. I could explain it in detail, show
code, but then, I'd be giving away features of my products that are
apparently 'unachievable', and no one is footing the bill for me to do this,
and I doubt you really care if I eat. For the most part, I will not be
including some of the cooler things I can do with Agent in my products,
because these features are of no use to an end user. But it can all be done,
without Microsoft lifting a finger, just not by sitting in a newsgroup and
ignoring possibilities.
I've been working with Agent for about a year. I read everyone's '3.0 wish
list', and figured out that it would be simple to accomplish most of what
was there with current technology. Every time someone on this group says
'Agent won't let me do this!!!', I get it to work. Maybe it's because I
believe it can be done and others would rather accept failure. Frankly, I
see failure as opportunity for improvement, and complaining as an
ineffectual waste of time.
If 3.0 uses the current format and fixes the scaling, that's good enough for
me. Any radical modification might destroy marketing possibilities AND
technological innovations available with the current version. If they
decided to split into two technologies, that would be the best way, so the
current sprite technology can do what you can with 2D and bitmaps and other
technology can benefit from advancements.
>How do you create a 3D background and have the agents "work" within it
>effectively? I don't think you can, not with a sprite-based engine.
Hence my statements that there are other technologies for accomplishing
these tasks. If you don't like the sprite engine, you mention plenty of
alternatives, use them. Sprites are relatively cheap resource wise. I like
to use a lot of Agent Characters. Huge 3D this that or the others that slow
down my web experience get deleted. Also, the processor usage would
interfere with speech processing.
I'd love to see 'the greatest thing', but I actually like to use my computer
for other things while there's an Agent Character on screen. I'm assuming
the developers of Agent have this in mind as well. I like Anfy 3D and other
tools I've seen, but the unique properties of Agent make it more useful than
any other technology. Also, you can use these other technologies in ADDITION
to Agent to accomplish your design goals.
If you really want to make full screen interactive presentations, look into
the technology behind an old video game, Dragon's Lair. It used a branching
methodology with LaserDisc technology to accomplish full screen
interactivity. The same theories could be used with a CD-ROM today. You
could make anything you could imagine and it would cost $1 per copy to
reproduce. Record your whole presentation into full screen video and there
you go, you can slap anything you want in there.
Anything huge over the web is ignored by normal people with their 28.8's
33.6's and 56K modems.
>However, Agent has been used beyond its original design
>specifications. At I Pyxidis we were using them as virtual actors
>performing mostly scripted lines in realtime, similar to how Pulse3D
>uses its characters. These features would have been very useful for
>that.
I have yet to see a single character file, application, or demonstration
from I Pyxidis, so I have no idea of the usefulness of what was developed
there. There might have been a lot of great things done there, but
apparently no one can experience them. I'd be more than happy to evaluate
this software and it's use of Agent but in my reality it simply doesn't
exist because I was never able to download it or look at it. If they put it
up on the web and it's enthralling and offers new possibilities, there's
something to be said, but it's much ado about nothing unless I can play with
it.
Rob
Rob Lindman
Glenn Saunders wrote>
>That's the problem. Agent's conception was too narrow, and hence its
>usefulness was curtailed by the fact that it _isn't_ a generalizable
>display engine.
You aren't going to get anywhere by attempting to claim Agent isn't useful.
It's usefulness to your particular vision might be curtailed by your desire
to see it as a 'generalizable display engine', but Agent isn't a display
engine at all. It's a user interface enhancement. If you want to do huge
things with realtime 3D, go get the code to Quake. If you want speech in it,
add code for Speech API. It's that simple.
>While Agent really isn't that useful in writing games (or creating
>scripted virtual characters that interact with eachother and
>background), GAME engines are up to the task of being applied to
>creating AGENT UIs, which is all MS Agent really is.
Then do it. I think Agent is, while in need of some upgrades, perfect for
the uses set forward by it's creators.
>
>Game engines encompass the requirements of an Agent UI and go far
>beyond them! So you see, it would not be a bad thing for a next-gen
>Agent control to be done this way--it would actually give it more of a
>chance of being a success as it would be useful in so many arenas, not
>just the strict notion of MS Agent.
>So making MS Agent more into a generalizable display engine, it
>retains its usfeulness as a UI, but also opens up all sorts of other
>applications in one fell swoop.
No. Agent's success isn't based on it become some huge all encompasing
gaming technology. You've just chosen to believe that. Your belief is
incorrect. Technology for other uses you suggest exists, you've just ignored
it because you don't want to expand beyond Agent development. Other people
will, and you'll be in the dust.
>Let me give you an example. Using MS Agent in multiuser chat.
>
>You get all the agents on a web page talking. First of all, no
>overlapping dialogue possible. That kinda sucks.
To a certain extent, I agree about asynchronous sound/speech, but at the
same time, I'd also be annoyed greatly by this. Most chat rooms are full of
crap. Do I really want OVERLAPPING crap? I'd rather set my computer on fire
than hear ALL of the crap in an IRC channel simultaneously. It would be too
much to deal with.
Secondly, all the
>characters are 3D rendered sprites all perspective-flat across the
>entire screen. You can not impose any sort of workable 3D background
>for them to be in for chat-room atmosphere.
That's your theory, not mine. Just because Agent doesn't have this feature
built in doesn't mean it can't be written. Also, not something inherently
necessary to the current features and benefits of Agent technology.
>Maybe you can do a Diablo sort of thing with Isometric, but it's still
pretty lame. You can't have the characters pick up background elements.
>It's a pure backdrop, noninteractive.
Bah. 'I've fallen, and I can't get up.' 'Mommy, I can't do it'.
>You can't really have characters walk from
>place to place. They "glide" with movetos. Basically they just look
>like they are hovering over the page, not a part of it.
There's ways to move Agent and make it look like it's walking. Not GREAT
ones, but it's still possible to improve on what's been done by other
developers, if you make it your mission to do so.
One of the first things I ever published to this group was my 'walking
engine', which i called 'MSopotamia'. I used VB to handle the walking
animation of an independent sprite. I could add an Agent Character to that
and it would be an improvement on current movement methodologies. It's not
like it was all that wonderful, but it could handle the task. it was 16K. It
took me 5 minutes to write. And yet, by your definition, it can't be done
unless all of the usability of Agent and all of the research and development
by everyone are thrown away to make some huge thing that does everything so
Glenn can take a nap.
Also, there's plenty of animation capability in HTML that you could swap
between an Agent and an HTML graphic if you wanted more out of Agent web
use, and interactive in the document.
>
>The TTS and the real-time animation make Agent really ideal for
>multiuser chat over other off-the-shelf plugins like straight VRML or
>the cumbersome and non-TTS Pluse-3D, but the fact that each character
>has no conception of their environment or what's going on with other
>characters hinders all sorts of applications like these.
>
First, go download the speech SDK, and merge a few controls from it with a
Pulse 3D or VRML app, because it's obvious you are in love with this stuff
although it has nothing to do with Grandma.
next, There are four great properties of Agent, .Left, .Top, .Height, and
.Width. You can use these to 'inform' characters where other characters are.
There's also the 'Request' Object, which lets characters know when other
characters are busy. What was it you were looking for here? The only
hinderance to creating applications you mention is the occupation of time
saying 'It can't be done.'
>What if you create a MASH script with two characters talking to
>eachother. First of all, unless the character has a suite of
>animations in 90' angles both left and right, they will appear to be
>in a play with them emoting straight ahead and not looking at
>eachother. Because it's bitmap based, the "download footprint" as you
>call it, would skyrocket to give the character enough positions to
>allow it to properly "emote" in all directions, sit down, roll around,
>and so on.
Because of my desire to teach myself software design rather than work to pay
for an education, I have only been able to afford to take 2 classes in
college. One of them was Theater. I was a spotlight operator. One thing I
noticed, on the stage, the actors were predominately facing the audience.
Sure they could 'emote' to each other with more angle and precision, but for
the most part the direction was towards the audience. And you aren't limited
in character design, so you can do whatever you want with your animations.
Also, effective programming of the Agent server can make downloading large
animations seamless. Make more animations, or use a better movie making
technology.
>MS Agent was designed so that its footprint would be manageable as
>long as the animations all branched off of a forward-neutral
>perspective. Once you want a truly generalizable character that can
>"emote" better than a block of wood you will be spending all your time
>rendering frames to create all the animations, and the end-user will
>be downloading animation all day.
New branching mechnisms would be cool. Also, adding more animations to an
existing ACF WHILE it is being viewed would be tremendous. However, if I
wanted a 'generalizable character', I'd use something else.
I saw these really cool 'superman 3D movies' on the web. That technology
would be far better for what you are talking about. You should check it out.
>Even a larger initial download for a realtime-3D agent engine would be
>made up MANY TIMES OVER in the flexibility of simply ordering the 3D
>character to turn 90' left or right and then firing off the other
>animations from the new perspective.
A larger initial download would destroy web viability of new markets for
Agent. The current download is just BARELY small enough that you might be
able to get people to use it. To market anything with Agent, the small
footprint is essential.
No larger initial download would be made up if people didn't download it in
the first place. If they leave my website because it's taking to long, quite
frankly, I'm an idiot for using it.
When bandwidth isn't an issue, I might be inclined to agree with file size
issues, but right now, no way.
> It makes sense to leverage game know-how for MS Agent.
I've basically drawn the conclusion that you've been thinking of Agent as an
end-all to creating animation and games on the web, and it's not, and I
don't know that it's intended as such. We're talking about an application
helper technology here.
It does make sense to leverage game know how to a certain extent. It also
makes sense to consider other technologies when creating animation and
games, which is something I believe you aren't doing. Rather than develop
for or explore other technologies, you'd rather try to mutate Agent into
something it simply is not, and probably will never be.
I don't think Agent is here to necessarily 'take over our screen and wow
us', it's here to be ubiquitous with the Windows interface and provide a
more natural interaction method than clicking on little squares and dialog
buttons.
Agent is here for Grandma. Maybe you don't realize that, but in every
article produced by Microsoft, they make it extremely apparent. They might
make it 3D, but I really doubt they are going to redo the whole GDI and call
it Agent 3.0.
Rob
Rob Lindman
> I'm guessing that the DOS box may soon be welcoming Genie home. . .
Bah. It was said Agent is dead a year ago. You dorks are still here. :P
Besides, you never know what could be here, tomorrow.
Rob
"What kind of dork are you today?"
T-man
I think I might make a soapbox (Mr Softsuds) agent with Rob's head sticking out of it. To move, a stagehand comes in from off-stage
and pushes him to a different spike. What's funniest (but no one can see it) is that he has eyes in the back of his head so when
something pops up on the backdrop, he knows which way to toss his head to indicate what the hell he's talking about. The spot-op
has the easiest gig, since he just locks off on the talking head. He falls asleep quite a bit because in spite of the quality of
the material, his delivery is only slightly better than a speak and spell, and Rob, true to MS form about such things, keeps falling
asleep, too.
T
"Rob Lindman" <thegre...@hotmail.com> wrote in message news:ODFmcuZr$GA.89@cppssbbsa03...
Phil Jackson
viable as it stands.
If you had to seriously invest in developing commercial software, you would
want to be confident that it's not going to be made unpredictably redundant
according to a regular upgrade cycle of Agent technology.
I'm quite impressed with what MS has done by providing free development
tools that have stimulated a variety of 3rd party developers to profit from
the current technology.
Sure, I too would like to hear a better TTS, have a true A-I behind
characters and a less bandwidth for the files, but it's not as if the public
has embraced the concept yet any way. What commercial encouragement is MS
getting to take it further? None! There's no comparable alternative and
hence no competition (as far as I know).
It seems to me that MsAgent simply needs to evolve at a different pace than
browsers or OS's in accordance with the greater lag in being adopted by
users. I reckon Windows Millenium & 2000 will have to out number pevious MS
installed systems before changing Agent can be of effective group benefit.
.....Just my 2 cents worth.
- Phil Jackson.
Rob Lindman
I like your approach here...
Where do you find information that Merlin is more popular than Barney?
Rob
Stephanie Heck wrote in message ...
>We agree that certain upgrades to MS Agent would be nice. We disagree on
the
>nature of the additions. Prehaps we should compile a specific wish list ,
so
>that Tandy Trower can evaluate what he deems to be practical?
>
>I remember when the Beta edition of MS Agent was released. It was released
>by the Microsoft Corp. as a speech enabled enhanced GUI. MS Agent is a
>technology that is intended to enhance other technologies, It is only a end
>all solution for Internet cartoonists. I regret to say that your average
end
>user does not know the difference between A-I and GUI. Be advised that the
>MS Agnt character Merlin is more popular than Barney amoung pre schoolers.
>
>The potential of MS Agent has barely been touched.
>
>Cheers
>Stephanie Heck
>webm...@heck.com
>Message boards and voice / text chat http://www.sheck.com
>
>MS AGENT PAGE http://www.sheck.com/about.htm
>
>
>
>
>
>
>
Stephanie Heck
Stephanie Heck
Michael Holdsworth
SAPI 5 is released.news:u#6C6NZr$GA.256@cppssbbsa03...
Stephanie Heck
I took my laptop to a kindergarden, plugged in an auxillary speaker, and had
our old pal Merlin read stories to 3 classrooms of kindergarden kids.
The kids saw Merlin as the coolest thing since sliced cake. The kids were
given the choice of Barney, or more Merlin. Merlin won by 2 to 1. Barney is
just another tv character. Computerized characters are seen by the kids as
more sophisticated. Sort of upscale kewl :-) That is why computer games sell
on such a grand scale.
The next time I do this, I am going to use one of the female MS Agent
characters. The message I want to give little girls is that you dont have to
be male to be good at computers. The Microsoft Character Peedy has higher
ratings with little girls, and Merlin does better with boys. Just my
empirical experience that kids are facinated by MS Agent characters. Warner
Bros. already knows this. Cantouche has the contract. Congrats to Cantouche
for seeing economics as an intragal element of high technology.
More research needs to be done to see how kids react to MS Agent
characters. It has been my empirical experience that this MS Agent has
fabulous potential with very young kids. It is just a matter of research and
targeted marketing. MS Agent is far from dead. When speech engine technology
becomes better and more natural sounding, we will be using future MS Agent
characters for speech and elocution training. This MS Agent technology has a
huge future in multiple markets, get used to it. The best has yet to come.
When we started building MS Agent web pages, we never figured on being seen
as internet cartoonists by kids. Oh Well....anything legal for a price. In
short, we need to find a way to make this pay off for Microsoft and
ourselves as well. Yes, MS Agent needs a few technical updates but the
concept that it was founded on was ten years ahead of its time. You must
remember that it has been about 4 years since MS Agent Beta. Four years
after the beta was released, it is still seen as advanced GUI. What other
technology has held a sharp edge for that long? The Real Audio people
marketed a product called Seventh Level Agent as a fourth rate attempt to
imitate MS Agent and the product died in three months. So much for
competition. The MS Agent product was clearly superior.
We need to discuss this as a group, on a friendly level and compile a "wish
list" that will give Microsoft a pragmatic insight as to where the
developers stand in relation to the end user. Mr. Trower, and his team can
then determine what is practical from the supply end. We need to inform
Microsoft of where we can best promote the product. Certain members of this
forum have not always done that with the greatest maturity. We are human.
In an ideal marketing scenario, the developers and software company have a
reciprocation of marketing data to the benefit of the end user. The end user
is what it is all about. If we get behind the product, the end users will
get behind us. We will not be poor as a result. Newton had a valid point
about opposite and equal reactions. We might do well to ponder that as a
matter of economics.
I have to get to work now. I hope this explains my position better. I would
like to hear comments from Microsoft about the things I have spoken of here.
Stephanie Heck
webm...@sheck.com
T-man
Great stuff! I will never forget how my three year old reacted when I built a character out of one of his toy trucks. He
immediately understood it wasn't just more TV, and after a gear-whirring hesitation, he started talking back. . .blew me away. The
fact that most people who cannot read can speak and understand spoken language gives MSAgent a specific edge in applications
targeting illiteracy. The voice capabilities really cut to the quick.
T
"Stephanie Heck" <sh...@softhome.net> wrote in message news:OIL1rx#r$GA.284@cppssbbsa03...
Glenn Saunders
<thegre...@hotmail.com> wrote:
>You aren't going to get anywhere by attempting to claim Agent isn't useful.
I didn't say it wasn't useful. I said it isn't as useful as it could
be.
>It's usefulness to your particular vision might be curtailed by your desire
>to see it as a 'generalizable display engine', but Agent isn't a display
>engine at all. It's a user interface enhancement. If you want to do huge
What keeps UI's from being realtime 3D?
>Then do it. I think Agent is, while in need of some upgrades, perfect for
>the uses set forward by it's creators.
Within the framework of a proposal from almost five years ago,
perhaps. Don't you think there is a need to re-evaluate the
requirements?
>No. Agent's success isn't based on it become some huge all encompasing
>gaming technology. You've just chosen to believe that. Your belief is
Actually, if you look at MFC and DirectX, you will see that "general
purpose display engine" is BUILT INTO THE OPERATING SYSTEM and is used
by the windowing system (the UI) as well as games.
The next step is 3D. Microsoft themselves were trying this with the
CHROME project, remember?
>To a certain extent, I agree about asynchronous sound/speech, but at the
>same time, I'd also be annoyed greatly by this. Most chat rooms are full of
>crap. Do I really want OVERLAPPING crap? I'd rather set my computer on fire
>than hear ALL of the crap in an IRC channel simultaneously. It would be too
>much to deal with.
But not being able to do this is the problem. Putting limits on it
should be a choice, not an inevitability.
The problem with having to queue speech in a chatroom vs. IRC is
because what if everyone's speaking very long sentences? You spend
all your time waiting for the characters speaking to catch up with the
lines chatters are typing.
>That's your theory, not mine. Just because Agent doesn't have this feature
>built in doesn't mean it can't be written. Also, not something inherently
>necessary to the current features and benefits of Agent technology.
If it's written, it's pretty much a whole new program. It can not be
retrofitted into the current software, not by us at least.
>Bah. 'I've fallen, and I can't get up.' 'Mommy, I can't do it'.
Don't resort to name calling. I've worked against Agent's limitations
for over two years. I'm not speaking without experience.
>There's ways to move Agent and make it look like it's walking. Not GREAT
>ones, but it's still possible to improve on what's been done by other
>developers, if you make it your mission to do so.
I know, and these solutions still suck.
>took me 5 minutes to write. And yet, by your definition, it can't be done
>unless all of the usability of Agent and all of the research and development
>by everyone are thrown away to make some huge thing that does everything so
>Glenn can take a nap.
I'm telling you it sucks. The walk cycle occurs at a fixed rate. If
you keyframe the character moving around, it will be more like he's
swimming than really walking. I'm not talking about a HACK that
fulfills the minimum criteria for walking characters, I'm talking
about something that actually looks halfway decent.
>Also, there's plenty of animation capability in HTML that you could swap
>between an Agent and an HTML graphic if you wanted more out of Agent web
>use, and interactive in the document.
I've done that too. I've had characters appear to come out of stasis
tubes using DHTML filters and agent positioning routines. It's still
constraining.
>First, go download the speech SDK, and merge a few controls from it with a
>Pulse 3D or VRML app, because it's obvious you are in love with this stuff
>although it has nothing to do with Grandma.
Frankly, I don't give a shit about Grandma, and I don't know why
anyone else should have to either. Nobody's usage of a technology
should be constrained by the intentions of the creator. Many
wonderful applications would never have been written if people limited
themselves that way. Probably nobody would have ever written a game
for the original IBM PC, for instance. It was a "Business" computer,
after all. And there never would have been anything above CGA
graphics.
>next, There are four great properties of Agent, .Left, .Top, .Height, and
>.Width. You can use these to 'inform' characters where other characters are.
>There's also the 'Request' Object, which lets characters know when other
>characters are busy. What was it you were looking for here? The only
>hinderance to creating applications you mention is the occupation of time
>saying 'It can't be done.'
What about character Z-index? What about pixel-accurate collision
detection, not bounding box collisions?
>noticed, on the stage, the actors were predominately facing the audience.
This works fine with single characters, but not when multiple
characters are talking to eachother. It looks okay in a play because
it is an accepted convention, but it looks really stupid with an "MS
Agent" play. How often do you have long conversations with other
people without eye contact? In my case, I created glancing left and
right animations, but these would still leave the character's body
facing forward. If I created whole body rotate lefts and rights it
would bloat filesize.
>in character design, so you can do whatever you want with your animations.
But if you intend the internet as your delivery method, you are
limited in how large your characters can get.
The reason vector is popular is because it is quick to download.
Flash has millions of downloads because of this.
Defending MS Agent on the basis that you can create hundreds of
separate animations to cover any needed angle is stupid. It bloats
filesize and takes forever to render out, compile, and write out
ACF/ACSs.
>Also, effective programming of the Agent server can make downloading large
>animations seamless.
That depends on how the script is written. You can't get something
for nothing.
>A larger initial download would destroy web viability of new markets for
Initial download is already very large for the vast majority of web
sites which employ MS Agent because they force you to download entire
ACS files before you see a restpose.
>I've basically drawn the conclusion that you've been thinking of Agent as an
>end-all to creating animation and games on the web, and it's not, and I
>don't know that it's intended as such. We're talking about an application
>helper technology here.
It's application is whatever we want to use it for. It is efficient
to create a technology with as broad an application as possible,
ESPECIALLY CONSIDERING THE LACK OF USAGE OF MS AGENT up to this point.
I blame the scarcity of MS Agent applications on these limitations
discussed.
Now, there are some sites out there that use MS Agent similarly to
what I am discussing, chat and little virtual plays. You don't
criticize these sites as being a bastardization of MS Agent. Yet if
Agent were to be enhanced in ways which were to directly benefit these
two applications, which are, as you consider, NOT central to Agent's
true purpose, then suddenly that's a bad thing?
>Rather than develop
>for or explore other technologies, you'd rather try to mutate Agent into
>something it simply is not, and probably will never be.
I just don't see such a strong line between game technology and UI as
you do. Isn't GDI+ in Windows 2000 pretty much the process of wedging
DirectX-like faster graphics into the regular OS painting routines?
And while DirectX was designed with games in mind, it is also used for
things like TV tuner cards and the Windows Media Player. Game
technology is always on the cutting edge of graphics. For anything
designed with graphics in mind to purposefully take a blind eye to
game engines is foolish.
It's almost as if you look down on games as being not academic or
serious enough to be used in MS Agent. I'm sorry, but aside from
OpenGL, almost all innovative graphics initiatives have evolved out of
the need for better game graphics.
>I don't think Agent is here to necessarily 'take over our screen and wow
>us', it's here to be ubiquitous with the Windows interface and provide a
But there are a growing number of things in the Windows OS designed to
wow us. Window bar gradients, and wWhat about the new dissolve-in look
of the Windows 2000 UI?
Plenty of UI's are fancy. Metacreations UI's, or all those
nonrectangular UI's you see in Windows these days, WinAMP skins, and
so on.
And the new MAC UI has translucent windows.
>Agent is here for Grandma. Maybe you don't realize that, but in every
>article produced by Microsoft, they make it extremely apparent.
I don't see why "Grandma" wouldn't accept a 3D character where she
would accept a pre-rendered sprite.
Glenn Saunders
<thegre...@hotmail.com> wrote:
>Microsoft creates software components for developers to build upon. This
>creates a market for Microsofts developer tools. It's up to the developer to
>use the tools in their own innovative way to create markets for their own
>products or services. Would you prefer that Microsoft simply makes
>everything and therefore defeats the purpose of designing software entirely?
What I'm saying is that MS Agent has been hyped to be something out of
the box which it clearly is not. And so far I haven't seen a single
developer wrap MS Agent in such a way to MAKE it what it was hyped to
be.
That is not a criticism against developers, merely an observation.
There is much prideful chest-beating about how great and innovative MS
Agent is as "intelligent agent" technology which I think is misplaced.
>When Grandma resets it to 640x480 somehow, she can't get it back without
If she's smart enough to reset it, she's smart enough to fix it.
>A smart developer writes a program for Agent that says "Grandma, your
>computer can be changed so that your baby grandson's picture looks better.
Point me to a MS Agent application which displays this level of
intelligence.
>There are ways to control Z-index. Props can be accomplished by intelligent
>animation design. Characters aren't designed to be behind HTML elements,
>they are supposed to be in the foreground of all applications, but you could
>do this if you put some thought into it. I could explain it in detail, show
I know of a couple workarounds, and they are too labor-intensive to be
practical.
>including some of the cooler things I can do with Agent in my products,
>because these features are of no use to an end user. But it can all be done,
>without Microsoft lifting a finger, just not by sitting in a newsgroup and
>ignoring possibilities.
The conclusions I've made about Agent, as stated, were reached after
years of experience. I'm anxious to see your innovative solutions.
>If 3.0 uses the current format and fixes the scaling, that's good enough for
>me. Any radical modification might destroy marketing possibilities AND
>technological innovations available with the current version. If they
NOW we appear to have found the main bone of contention. You are
afraid that if a new Agent technology makes the current one obsolute,
your software development will have been wasted.
That's your bias. I'm looking for the best mousetrap, not protecting
an old mousetrap because I sell bait that only works with the old
ones.
But anyway, I feel that the two technologies can coexist, but let's
allow the best one to win, not stifle innovation out of fear of
change.
>to use a lot of Agent Characters. Huge 3D this that or the others that slow
>down my web experience get deleted. Also, the processor usage would
>interfere with speech processing.
A couple years ago, yes, but not with today's average computer
hadrware. The graphics chip handles the vast majority of the graphics
just fine without CPU intervention.
>If you really want to make full screen interactive presentations, look into
>the technology behind an old video game, Dragon's Lair. It used a branching
Stop assuming to know what I am looking for. Dragon's Lair requires
millions of dollars invested in traditional 2D ink and paint
animation, photography, film to video transfers, CD authoring and
distribution.
Realtime engines are scriptable with text editors (ala MASH) and
therefore much easier and cheaper to produce once you create your cast
of characters and some props.
>Anything huge over the web is ignored by normal people with their 28.8's
>33.6's and 56K modems.
You have no way to assume how "huge" it would be.
>this software and it's use of Agent but in my reality it simply doesn't
>exist because I was never able to download it or look at it. If they put it
I can't see any of your stuff either, which is apparently as you like
it.
Rob Lindman
> Nobody's usage of a technology
> should be constrained by the intentions of the creator.
Ok, here's my reply to that. I'd like to use my microwave to cook my kid
brother, but they put this damn door on it and I can't fit him in and still
close it. How dare they!?! Guess I'll have to use the stove instead. See the
'innovation' that took place? Sure, it will take longer to cook him, but he
will be juicier and more flavorful.
> >You aren't going to get anywhere by attempting to claim Agent isn't
useful.
>
> I didn't say it wasn't useful. I said it isn't as useful as it could
> be.
Ok. I'll grant you that. How useful CAN it be?
> What keeps UI's from being realtime 3D?
Grandma's 486 running Windows 95. If you looked at the MS Research site for
the 'Persona Project', it had a 3D Peedy, so obviously they've thought about
it, but like I've said, there are other uses for the 2D engine. Maybe they
want to do more with it first.
>Agent is here for Grandma. Maybe you don't realize that, but in every
> >article produced by Microsoft, they make it extremely apparent.
>
> I don't see why "Grandma" wouldn't accept a 3D character where she
> would accept a pre-rendered sprite.
Grandma wouldn't accept that a 3D character makes her computer run really
slow when she opens Microsoft Word to make a greeting card.
> Frankly, I don't give a shit about Grandma, and I don't know why
> anyone else should have to either.
Microsoft makes an average of 80 cents a day from every Grandma running
Windows, DOS, or anything else they make. If I had 80 cents a day from
everyone's Grandma, well, I wouldn't worry too much about geeks who want 3D
everything or a Java toaster until Grandma was taking a dirt nap.
> Within the framework of a proposal from almost five years ago,
> perhaps. Don't you think there is a need to re-evaluate the
> requirements?
I'd imagine they want to make Agent much cooler before upgrading. The 1.5 to
2.0 upgrade was probably to test the waters, see if anyone did anything with
it, and then go for higher purposes. I'm quite certain they re-evaluate
everything at Microsoft, every day.
> Actually, if you look at MFC and DirectX, you will see that "general
> purpose display engine" is BUILT INTO THE OPERATING SYSTEM and is used
> by the windowing system (the UI) as well as games.
>
Ok. So, you've proven my point. You could use DirectX and achieve what you
want, without destroying Agent's bitmap format. A juicier, more flavorful
meal, just by using the oven.
> The next step is 3D. Microsoft themselves were trying this with the
> CHROME project, remember?
CHROME was supposed to be some sort of a browser upgrade, and from what I
can tell, it's still in development. I think they've renamed it
ChromeEffects and it's part of the browser. There's a site on MS's website
somewhere about it. I know one guy in this group who could tell you more
about it than I could.
> >To a certain extent, I agree about asynchronous sound/speech...
>
> But not being able to do this is the problem. Putting limits on it
> should be a choice, not an inevitability.
It's not an inevitability. There's this great thing called the Speech SDK
which includes a sample of how to do this. It's without Agent, but you can
do it with Agent, you just have to work a little harder, or wait for
Microsoft. Do it first, maybe they will buy it from you. Wait for it, then
complain about how they've stolen your great idea and opportunity.
> The problem with having to queue speech in a chatroom vs. IRC is
> because what if everyone's speaking very long sentences? You spend
> all your time waiting for the characters speaking to catch up with the
> lines chatters are typing.
So in your theory it would be beneficial to hear all of those long sentences
as an unintelligeble mass of garbage, as long as it was over with in real
time?
> If it's written, it's pretty much a whole new program. It can not be
> retrofitted into the current software, not by us at least.
That's not necessarily the case, although it could be the case, too. You
could do both. As for 'US', you are the only one of the two of us that can't
do it. Agent isn't so complicated that some of it's key benefits can't be
duplicated.
> >Bah. 'I've fallen, and I can't get up.' 'Mommy, I can't do it'.
>
> Don't resort to name calling. I've worked against Agent's limitations
> for over two years. I'm not speaking without experience.
I don't believe you are without experience. I wouldn't have re-read all of
your newsgroup posts if I did. I'm just saying, I'm going to laugh at anyone
who gets on a podium and goes 'Why can't I do this?' when it seemed pretty
straightforward to me what you were looking to do. I've worked against
Agents limitations for 1 year, and my own for 26. Suffice it to say, I've
had more success with Agent's limitations than my own.
> The walk cycle occurs at a fixed rate. If
> you keyframe the character moving around, it will be more like he's
> swimming than really walking. I'm not talking about a HACK that
> fulfills the minimum criteria for walking characters, I'm talking
> about something that actually looks halfway decent
Ok. Not sure if you looked at my 'hack'. Again, you are suggesting Microsoft
pull something out of their collective ass that isn't necessarily easily
achieved. I've looked at the 250+ characters on the web, they all have
different feet, some don't even have them. Getting them to walk in a
believable fashion would be quite an accomplishment. Mathematically dividing
the distance between two points, calculating the number of footsteps, and
animating them without knowing how many feet it has would be difficult.
Defining the number of feet would exclude footless characters. What if I
made an Agent with 6 feet, like Ant, and one with no feet, like the paper
clip. Where am I going to get info to calculate the walking information for
a paperclip? Even Quake III has trouble with this, and it's advanced enough
by my reckoning.
> I've done that too. I've had characters appear to come out of stasis
> tubes using DHTML filters and agent positioning routines. It's still
> constraining.
Ok. Haven't seen that, sounds cool enough to me. Where's the website?
As for 'constraining', here, beat this 'limitation'. I want Peedy to jump
out of the screen, and land on my shoulder. Hows that for 'constraining'.
Damn Microsoft for not being able to figure that out, how dare they limit my
creativity. And then, instead of Peedy, I want a double cheeseburger. No
matter how you look at it, there will always be 'constraints'. Make an agent
out of a piece of fudge, and make it come out of my printer. I'll feel more
constrained than you until that happens, because I'm damn hungry, and I like
fudge.
>Many
> wonderful applications would never have been written if people limited
> themselves that way. Probably nobody would have ever written a game
> for the original IBM PC, for instance. It was a "Business" computer,
> after all. And there never would have been anything above CGA
> graphics.
Again, you are your own worst enemy here. I'm quite sure that if Agent has a
limitation, I could create a workaround. As far as CGA developers, well, you
again prove my point. People didn't accept the limitations imposed on them,
and new technology emerges. People who do accept the limitations, don't
create new technology. CGA developers had to accept only having 4 colors,
they still wrote games, and other guys came out with EGA, VGA, XGA, and
super VGA.
> What about character Z-index? What about pixel-accurate collision
> detection, not bounding box collisions?
As I said, Z-Index can be done. Pixel collision, while I hadn't thought
about it before, I have an idea of how it could... that one would be tough,
or cludgy, especially when animating.
> If I created whole body rotate lefts and rights it
> would bloat filesize.
Ok. You are right here.
> But if you intend the internet as your delivery method, you are
> limited in how large your characters can get.
Welcome to the exciting world of MODulator DEModulators.
>
> The reason vector is popular is because it is quick to download.
> Flash has millions of downloads because of this.
>
> Defending MS Agent on the basis that you can create hundreds of
> separate animations to cover any needed angle is stupid. It bloats
> filesize and takes forever to render out, compile, and write out
> ACF/ACSs.
Ok. I'm stupid. I can accept that, and try to learn how not to be.
Obviously, it's necessary to render something from every conceivable angle
on the desktop with Agent, when you could just resort to using Flash in the
browser.
> >Also, effective programming of the Agent server can make downloading
large
> >animations seamless.
>
> That depends on how the script is written. You can't get something
> for nothing.
Ok. How much did you pay for Microsoft Agent?
> >A larger initial download would destroy web viability of new markets for
>
> Initial download is already very large for the vast majority of web
> sites which employ MS Agent because they force you to download entire
> ACS files before you see a restpose.
That's not Microsoft's fault. More people should use ACF. I'd offer to
convert people's ACS files to ACF if they were willing to pay a small fee.
I've got the technology to do as much of it as you can conceivably do. I
offered this before, not one character developer emailed me, so I abandoned
development on that end and looked to more lucrative opportunities.
> It's application is whatever we want to use it for. It is efficient
> to create a technology with as broad an application as possible,
> ESPECIALLY CONSIDERING THE LACK OF USAGE OF MS AGENT up to this point.
Lack of Agent Usage? Ahem, MICROSOFT OFFICE. Maybe Bernard or someone from
MS can tell us how many copies of that baby are out there? If you mean small
developer projects that use Agent, I guess there aren't many, maybe it's
because only about 70 people have ever been to this group, and application
helpers aren't the 'hottest technology'?
However, the most profound Microsoft Agent application could be sitting on
some guy's computer, just waiting for him to have enough money for a printer
and a domain name. Maybe he's right across the street from Microsoft right
now, waiting for them to open on Monday and approve his request to
distribute Agent. :)
> I blame the scarcity of MS Agent applications on these limitations
> discussed.
Blame is a wonderful thing, isn't it. I blame you, for complaining about
what doesn't exist instead of just making more things with MS Agent. What
good does that do either of us? It's all your fault. It's all Microsoft's
fault. I say that it's because Emmanuel Lewis takes karate lessons.
>
> Now, there are some sites out there that use MS Agent similarly to
> what I am discussing, chat and little virtual plays. You don't
> criticize these sites as being a bastardization of MS Agent. Yet if
> Agent were to be enhanced in ways which were to directly benefit these
> two applications, which are, as you consider, NOT central to Agent's
> true purpose, then suddenly that's a bad thing?
I don't think anything is a 'bastardization' of Agent. I just don't think
they should throw the current functionality down the toilet just for you. Of
course, they might think the opposite, and think, "Why keep it the same just
for this Rob guy?" Personally, I think that if they had to start it off with
a derivative of my name, they should give me a shot at what I have to show
the world before messing it up completely.
> For anything
> designed with graphics in mind to purposefully take a blind eye to
> game engines is foolish.
It's not necessarily a given that EVERYTHING designed with graphics in mind
be used for gaming technology.
> It's almost as if you look down on games as being not academic or
> serious enough to be used in MS Agent. I'm sorry, but aside from
> OpenGL, almost all innovative graphics initiatives have evolved out of
> the need for better game graphics.
Ok. That's your theory. I don't think every graphics program was developed
JUST FOR GAMES. I think your gaming background influences you to believe
that. As for games, I like them very much. I'll frag your ass any time at
Quake III, you WILL become meat. But I also don't think Agent should be a
'Quake III' replacement, either. Academics are fine, I am doing this for
educational purposes, but a good ass kicking works in well once in a while
:)
> Plenty of UI's are fancy. Metacreations UI's, or all those
> nonrectangular UI's you see in Windows these days, WinAMP skins, and
> so on. And the new MAC UI has translucent windows.
IBM, UBM, we all BM. UI, my eye, bye bye. Go ahead, MAC my day.
Rob
Rob Lindman
> the box which it clearly is not. And so far I haven't seen a single
> developer wrap MS Agent in such a way to MAKE it what it was hyped to
> be.
I'll have a response to that soon enough. Also, maybe you haven't looked at
ALL of the MS Agent research sites Microsoft has had on the web. As for
thinking out of the box, I can't think of anything more out of the box than
what I'm about to show the world with Microsoft Agent.
> That is not a criticism against developers, merely an observation.
> There is much prideful chest-beating about how great and innovative MS
> Agent is as "intelligent agent" technology which I think is misplaced.
That may be true. And then, maybe like Bill keeps saying, "The best is yet
to come."
> >A smart developer writes a program for Agent that says "Grandma, your
> >computer can be changed so that your baby grandson's picture looks
better.
>
> Point me to a MS Agent application which displays this level of
> intelligence.
What am I here to do, hold your hand? For a guy who's spent twice as long on
this as I have, you'd think you would have written it by now. You've had the
money and a team to do it. I've had ME. It takes time to write things like
this. I'm good enough at it, but this isn't what I'm looking to do. I have
higher priorities. Microsoft will write that program, I'm sure.
> >There are ways to control Z-index. Props can be accomplished by
intelligent
> >animation design.
>
> I know of a couple workarounds, and they are too labor-intensive to be
> practical.
Use another technology rather than wait for MS to reinvent Agent if you want
to be practical for your specific purposes. It would be simple enough to do
in the timeframe you've been here.
> The conclusions I've made about Agent, as stated, were reached after
> years of experience. I'm anxious to see your innovative solutions.
You won't be seeing any of them for quite some time, you'll see my
marketable solutions, if I decide to announce them in this group, or if you
come across them on the web. I've looked at Agent in every way imaginable, I
know what the 'holy grail' is, I know I can make it.
However, Albert Einstein has a marvelous quote, "Out of Clutter, find
Simplicity.
From Discord, find harmony. In the middle of difficulty, lies opportunity."
I live by this quote. What I will put on the web will be simple, and from
there, I will expand into more complex themes which require 'innovation'.
> >If 3.0 uses the current format and fixes the scaling, that's good enough
for
> >me. Any radical modification might destroy marketing possibilities AND
> >technological innovations available with the current version. If they
>
> NOW we appear to have found the main bone of contention. You are
> afraid that if a new Agent technology makes the current one obsolute,
> your software development will have been wasted.
I am sufficiently prepared to rapidly adapt to a new version of Agent should
one appear within the next 24 hours, 24 days, or 24 years.
> That's your bias. I'm looking for the best mousetrap, not protecting
> an old mousetrap because I sell bait that only works with the old
> ones.
No, you stopped looking for the best mousetrap, and sat in this group and
whined about why this one isn't it. Of course I don't want my software to be
obsolete, but that's hardly something I have to be concerned about here.
Also, I've seen the best mousetrap (yes, an actual mousetrap) ever invented,
and it was a marketing failure. Perhaps when I find time I will post a
diagram of it and explain why no one bought it, or maybe it's already on the
web somewhere.
> But anyway, I feel that the two technologies can coexist, but let's
> allow the best one to win, not stifle innovation out of fear of
> change.
If your idea of innovation is abandoning an existing mousetrap to build an
entirely new one, then your idea is incorrect. Take the best existing
mousetrap, and add an electric fence.
What you propose is not an innovation in this area, but an abandonment of
current advantages for a replacement system. I'm quite certain what your
bias is, and I have a good idea what's motivating it. Still, your
perspective is limited to your usage, and I am attempting to understand
everyone's usage, Microsoft's included. You want a game engine, Microsoft
wants an application helper and conversational user interface that's non
intrusive. We differ on our viewpoint, but I am trying to defend the
generalized nature of Agent, and you ignore the existance of other
technologies rather
> Stop assuming to know what I am looking for.
Stop assuming that your view of what Agent should be is more valuable than
the time and effort of the creators of the existing 250 characters.
> Dragon's Lair requires
> millions of dollars invested in traditional 2D ink and paint
> animation, photography, film to video transfers, CD authoring and
> distribution.
> Realtime engines are scriptable with text editors (ala MASH) and
> therefore much easier and cheaper to produce once you create your cast
> of characters and some props.
I'd imagine the same thing could be achieved with Flash today in a much more
affordable manner. Again, use another technology if you don't like the
limitations of the one.
> I can't see any of your stuff either, which is apparently as you like
> it.
I haven't put anything out on the web because I haven't had a cent to spare
to get a website, I haven't had the money for a lawyer to get trademarks and
copyrights, but I do now.
I've had to learn 10 times more than anyone else to get a product to the
web, and work 10 times as hard as the GROUPS of other developers. And I am
ready.
I've worked on Agent while I lived in my car, I've written code on napkins,
toilet paper, logged on to this group from the public library when I was too
poor to afford a computer. My life has never been easy. Maybe that's why I
don't accept the same limitations you do, because if I did, I would be dead.
But I am here now, and my guns are blazing, and I know that I am right on
target.
Whether or not you believe it is inconsequential, you are just a little
rectangle I like to argue with. You've had 10 times the resources and
experience I've had, and you've got nothing to show for it. I haven't even
started.
That's why this trip is important to me, and why I've told everyone here
about it. I was lucky enough to get a donation for the trip, so I could see
Microsoft, so I could understand, so I could get my shot at what you have
apparently taken for granted, the opportunity presented by the simplicity in
all of the clutter.
But rest assured, I wasn't satisfied with being unable to put Agent in the
microwave, so I did put it in the oven. And it WILL BE juicier and more
flavorful.
End Of Line.
Glenn Saunders
<thegre...@hotmail.com> wrote:
>Ok, here's my reply to that. I'd like to use my microwave to cook my kid
>brother, but they put this damn door on it and I can't fit him in and still
>close it. How dare they!?! Guess I'll have to use the stove instead. See the
>'innovation' that took place? Sure, it will take longer to cook him, but he
>will be juicier and more flavorful.
That is an absurd analogy. There are countless examples of computer
applications being used in positive creative ways.
>Ok. I'll grant you that. How useful CAN it be?
Already been stated.
>Grandma's 486 running Windows 95.
486s can't run MS Agent very well. Pentium 200s have a little trouble
with them as it is.
>Grandma wouldn't accept that a 3D character makes her computer run really
>slow when she opens Microsoft Word to make a greeting card.
You have no evidence to show that it would slow her computer to a
crawl.
>Microsoft makes an average of 80 cents a day from every Grandma running
Evidence?
>It's not an inevitability. There's this great thing called the Speech SDK
>which includes a sample of how to do this. It's without Agent, but you can
I think any endeavor to create a new Agent technology outside of
Microsoft should be cross-platform. Limiting Agent just to Windows is
another reason it isn't as popular as it could be.
Making it cross platform involves jumping through more technological
hurdles, but it's worth it.
>So in your theory it would be beneficial to hear all of those long sentences
>as an unintelligeble mass of garbage, as long as it was over with in real
>time?
How often do you enter into a real chat and wind up with an
unintelligeble mass of garbage? People will moderate the timing of
their speech to accomodate this. There is always a way to strike a
balance. Nobody wants to enter a chat room and contribute to
cacophany. It defeats the purpose.
>Ok. Not sure if you looked at my 'hack'. Again, you are suggesting Microsoft
>pull something out of their collective ass that isn't necessarily easily
>achieved. I've looked at the 250+ characters on the web, they all have
>different feet, some don't even have them. Getting them to walk in a
At least in sprites, this is a problem solved ages ago in game
development. The rate at which a character walks should influence the
pace of the animation playback of the walk cycle. This should stop
foot-drag or other problems.
>Ok. Haven't seen that, sounds cool enough to me. Where's the website?
I did the work on salary for the startup company which is now in
limbo. I was not allowed to share any of this work. The stills you
see on my home page are about all I'm at liberty to share.
http://www.geocities.com/Hollywood/1698
When the company was spiralling down I tried to convince them to
release some of the characters for non-profit use. They agreed in
principle but never actually went through with it. So none of the
characters are out there.
I have attempted to solicit others to purchase I Pyxidis' intellectual
property but nobody seems to be that serious about putting up some
bucks. I don't think it would cost that much, but then again, there
doesn't seem to be that much $ to be made with MS Agent.
>Again, you are your own worst enemy here. I'm quite sure that if Agent has a
>limitation, I could create a workaround. As far as CGA developers, well, you
>again prove my point. People didn't accept the limitations imposed on them,
>and new technology emerges. People who do accept the limitations, don't
>create new technology. CGA developers had to accept only having 4 colors,
>they still wrote games, and other guys came out with EGA, VGA, XGA, and
>super VGA.
So your only issue is that I'm expecting Microsoft to release Agent 3
instead of me writing it from scratch? Come on. Not everyone has the
resources to pull off projects of that magnitude. If you do, I'll be
the first to give you credit when you release your software.
People who make software development requests are not whiners just
because they aren't qualified to write it themselves. Microsoft has
almost infinite resources to bear. I read in the monopoly filings
that they had over 1000 developers creating Internet Explorer 4. This
is a company that can and should do a lot more with MS Agent than what
we have seen.
Responding with a "do it yourself" is just plain silly. There is no
equivalence between Microsoft's capabilities to develop Agent
technology and the capabilities of any of the independent developers
on this newsgroup.
For the most part, we are USERS of MS Agent technology. We create
characters, we wrap the control into useful applications, build
websites that invoke the agent as a control. We really aren't capable
and/or willing to do much more than that.
>As I said, Z-Index can be done.
How? The agent comes to the front anytime it animates or speaks,
right? You can't control how agents layer on top of eachother.
>Ok. I'm stupid. I can accept that, and try to learn how not to be.
>Obviously, it's necessary to render something from every conceivable angle
>on the desktop with Agent, when you could just resort to using Flash in the
>browser.
Flash doesn't support TTS. It's mostly a design-time technology, not
a runtime technology like MS Agent.
>Ok. How much did you pay for Microsoft Agent?
I didn't pay anything for IE either. I guess Microsoft is under no
obligation to further enhance its HTML rendering engine, then.
>That's not Microsoft's fault.
But if despite this people ARE willing to download ACSs, then this
makes your point moot since people would in turn likely to be willing
to sit and wait for a larger ActiveX control to install.
>Lack of Agent Usage? Ahem, MICROSOFT OFFICE. Maybe Bernard or someone from
I didn't say no Agent usage. I said lack of Agent usage. Microsoft
Office is a biggy, but still it's just a handful of programs using it
as a help system and they don't even let those characters speak or use
speech recognition. That's not really pushing the technology in ways
I think MS Agent can be most exciting, IMHO.
>IBM, UBM, we all BM. UI, my eye, bye bye. Go ahead, MAC my day.
Well, that was intelligent.
Glenn Saunders
<thegre...@hotmail.com> wrote:
>What am I here to do, hold your hand? For a guy who's spent twice as long on
Don't use your development endeavors in your arguments if you aren't
willing to elaborate. I've said as much as I can about what I did,
why can't you?
>this as I have, you'd think you would have written it by now. You've had the
>money and a team to do it. I've had ME. It takes time to write things like
>this. I'm good enough at it, but this isn't what I'm looking to do. I have
>higher priorities. Microsoft will write that program, I'm sure.
Actually I don't. I worked as part of a team for a startup company
that is no more. I created and scripted for the agents. Others wrote
wrappers around the control, chat clients and servers, and so on. We
were limited by the direction given to us by our boss. We could have
accomplished a lot more had the company focused more intently on MS
Agent for MS Agent's sake instead of their lofty goals of interactive
television, but they chose to look at MS Agent as a stopgap substitute
for future technology. Consequently, they missed out on some
opportunities to be clear leaders in MS Agent development, more
specifically with their chat client/server.
However, the experience was enough to show me what the potential was.
I am in no personal position to rewrite what was written at I Pyxidis,
nor to buy out their I.P. myself.
I would like to create some more characters and make them freely
distributable, and perhaps find either a vehicle to purchase I
Pyxidis' I.P. or secure funding to write a new chat server, but that's
dependent on whether my dayjob and my professional connections
provides such an opportunity. I certainly don't have the time nor
inclination to pursue MS Agent development fulltime outside of my
dayjob.
>Use another technology
There is no other technology. There could be, but there isn't. And
that's what's the most frustrating. You'd have to license some tech
here, some tech there, and glue it together as something new, and I'm
in no position to do that. I've researched plenty including just
about every 3D web technology out there. None of them have the right
combination of features to provide a substitute for MS Agent out of
the box. And there was no way to glue technologies together together
to hack it to work, either.
Some of the stuff I've looked into include: Alice3D, VRML/3DML,
Pulse3D, Director, Flash, and Flatland. They all have annoying
limitations of one sort or another.
I wish some entrepreneur would step in and fill that void, but so far
it hasn't happened. There is still this wide gap between developers
who write fancy 3D plugins for web browsers and what those things can
do, and the kinds of awesome graphics you see in PC games. If you can
merge the two, you've really got something. Plenty of PC games
network through TCP/IP. Some games like Everquest et. al. are only
playable with an internet connection. So the lines are blurred. It
must be that nobody wants to just give away their technology as a
general purpose 3D construction set. They want to keep it
proprietary. So we're left with cumbersome beasts like 3DML.
>there, I will expand into more complex themes which require 'innovation'.
I can't wait.
>If your idea of innovation is abandoning an existing mousetrap to build an
I never said people should actively abandon MS Agent. Old tech never
dies, it fades away. Some people still use DOS and Windows 3.1, for
instance. I still use an Amiga system.
MS Agent will succeed or fail based on its inherent merits and will
continue to be used to some degree even in lieu of "better"
technology.
Arguing that we should prevent newer rivals to MS Agent from coming
out to protect MS Agent from obsolescence is the mistake.
>everyone's usage, Microsoft's included. You want a game engine, Microsoft
A game engine is a superset of a UI that would not prevent MS Agent
3.0 from being used in a more restricted way. You could have 3D
merlins with no opaque background, small, in the corner of the screen
facing forward and firing off robotic animatronic-style moves as he
currently does if you wanted. But then you would have the option of
opening up a more elaborate 3D stage and have characters interact in a
more realistic fashion than is currently possible.
Have you ever played with one of the Catz or Dogz games? Very similar
to MS Agent. I've actually met the author. Those things rely on a
foundational graphic engine, the ballz engine. Can't you see the
overlap here? It's not apples and oranges.
>Stop assuming that your view of what Agent should be is more valuable than
>the time and effort of the creators of the existing 250 characters.
When did I say that?
>I'd imagine the same thing could be achieved with Flash today in a much more
>affordable manner. Again, use another technology if you don't like the
>limitations of the one.
Actually, while Flash streamlines the development cycle, it is still a
design-time engine, not a runtime engine. Flash4 goes to great
strides towards making things more scriptable and interactive, but
it's still not quite there. As previously discussed, I've looked at
both Director and Flash as substitute agents. Director, if for no
other reason, is insufficient because of its opaque background. Flash
transparency only works in IE. Neither can acheive an overlay effect
independent of application window as MS Agent can.
>I've had to learn 10 times more than anyone else to get a product to the
>web, and work 10 times as hard as the GROUPS of other developers. And I am
>ready.
Hat's off to you.
>I've worked on Agent while I lived in my car, I've written code on napkins,
You are the one taking this personally, and I can see why. I do
respect you, but I just disagree with your resistance against newer,
different types of agent technology, and the imporance you place on
the "barely computer literate with old PC running 16 colors"
demographic as Agent's target demographic.
Your dedication bests me, I'll give you that. I just hope for your
sake it pays off. MS Agent has been a glorified hobby for me, not
intended to be my bread and butter. I'd guess that is also the case
for most MS Agent developers.
I have not gambled my career and livelihood on MS Agent. I've become
a more general purpose web developer/Flash guy, which is a much lower
risk career strategy.
>That's why this trip is important to me, and why I've told everyone here
What's this trip all about?
>End Of Line.
Is that a Tron reference?
Rob Lindman
> Sure, it will take longer to cook him, but he
> >will be juicier and more flavorful.
>
> That is an absurd analogy. There are countless examples of computer
> applications being used in positive creative ways.
>
Is it really? I wanted a way to make agent more useful, I couldn't wait for
Microsoft, so I invented various ways to do it. However, I will admit, I do
love to be absurd. :)
> So your only issue is that I'm expecting Microsoft to release Agent 3
> instead of me writing it from scratch? Come on. Not everyone has the
> resources to pull off projects of that magnitude. If you do, I'll be
> the first to give you credit when you release your software.
Not everyone. Especially not me. But, I could probably duplicate the
majority of Agent 2's features which I find most useful in less than a week,
perhaps the whole thing in a month, without a lot of the limitations you
perceive, but why? The only thing which would present a problem would be the
control/class ID download, I couldn't afford a Verisign ID. I would also
need to use the L&H TTS engine, although I've got a couple of solutions
which are close in that area too. The code to Quake, which is available
online, would be a sufficient starting point to achieve all of the 3D
interactivity you've requested.
The internet provides everyone with intellectual property resources
comparable to those Microsoft has.
My issues against format destruction and are: the 250 characters which have
been developed under the bitmap format, the existing software that's on the
market which uses the bitmap characters, the unrealized benefits of the
current format, and the slower machines.
I am not against API improvements or enhancements. However, if they did
include a bunch of AI equivalents in their apps and if they did revamp the
entire thing, I'm sure there would be more people than you making a stink
about it.
> People who make software development requests are not whiners just
> because they aren't qualified to write it themselves. Microsoft has
> almost infinite resources to bear. I read in the monopoly filings
> that they had over 1000 developers creating Internet Explorer 4. This
> is a company that can and should do a lot more with MS Agent than what
> we have seen.
Ok. They've chosen not to, and you want more from the technology. You have
two choices, accept that they've chosen not to and build with what's there,
or make your own. You can try to become more qualified, or accept that you
aren't and wait.
> Responding with a "do it yourself" is just plain silly. There is no
> equivalence between Microsoft's capabilities to develop Agent
> technology and the capabilities of any of the independent developers
> on this newsgroup.
I'll never claim to be anything other than just plain silly :) There are
equivalences between software development at MS and development on the net.
Aside from their knowledge of Windows, the PC does present a level playing
field to all developers. Yes, that's a somewhat large aside, but more
information about Windows internals is released every day to the net.
>
> For the most part, we are USERS of MS Agent technology. We create
> characters, we wrap the control into useful applications, build
> websites that invoke the agent as a control. We really aren't capable
> and/or willing to do much more than that.
Ok, so the issue here is one USER verses one USER, so we negate each others
vote, and therefore, we are at a standstill. As for being capable or willing
to do more than that, I think I am capable of doing more, and I'm willing to
do whatever it takes to make my products better.
>
> >As I said, Z-Index can be done.
>
> How? The agent comes to the front anytime it animates or speaks,
> right? You can't control how agents layer on top of eachother.
I can't universally say 'this agent will always be on top? Hmm, that's a
suprise to me. I think there's a thing called the Windows GDI32.DLL file
that has API's in there. Yeah, it's not a web compatible solution, but you
can read up on it all you want.
>
> >Ok. I'm stupid. I can accept that, and try to learn how not to be.
> >Obviously, it's necessary to render something from every conceivable
angle
> >on the desktop with Agent, when you could just resort to using Flash in
the
> >browser.
>
> Flash doesn't support TTS. It's mostly a design-time technology, not
> a runtime technology like MS Agent.
Ok.
>
> >Ok. How much did you pay for Microsoft Agent?
>
> I didn't pay anything for IE either. I guess Microsoft is under no
> obligation to further enhance its HTML rendering engine, then.
>
> >That's not Microsoft's fault.
>
> But if despite this people ARE willing to download ACSs, then this
> makes your point moot since people would in turn likely to be willing
> to sit and wait for a larger ActiveX control to install.
You need to understand what is marketable and what is not. A large quantity
of users on the net DON'T have Agent, and WON'T download ACS files, and
don't know where WINDOWS/MSAGENT/CHARS is or how to copy a file there. A
large quantity of users stand to benefit from well designed applications
which use functions well within the reach of Agent 2.0.
> >Lack of Agent Usage? Ahem, MICROSOFT OFFICE. Maybe Bernard or someone
from
>
> I didn't say no Agent usage. I said lack of Agent usage. Microsoft
> Office is a biggy, but still it's just a handful of programs using it
> as a help system and they don't even let those characters speak or use
> speech recognition. That's not really pushing the technology in ways
> I think MS Agent can be most exciting, IMHO.
Ok.
> >IBM, UBM, we all BM. UI, my eye, bye bye. Go ahead, MAC my day.
>
> Well, that was intelligent.
>
I haven't gone out of my way to call you stupid, yet you've done so to me on
3 occasions now, and you're the one who said 'let's not resort to name
calling'. It will only serve to amuse me that you'll call me stupid when
it's now obvious I have obtained answers to problems you haven't solved, in
less time, and with less resources and experience.
Rob
Rob Lindman
> On Mon, 1 May 2000 03:10:10 -0700, "Rob Lindman"
> <thegre...@hotmail.com> wrote:
> >What am I here to do, hold your hand? For a guy who's spent twice as long
on
>
> Don't use your development endeavors in your arguments if you aren't
> willing to elaborate. I've said as much as I can about what I did,
> why can't you?
I can do whatever I want. I don't have to tell you anything. You can't
command me or dictate the manner in which I argue. I'm a bastard, I can get
away with it. I've shown you as much of my software as I've seen of yours,
none. I'd say that's fair and equal treatment.
>
> >this as I have, you'd think you would have written it by now. You've had
the
> >money and a team to do it. I've had ME. It takes time to write things
like
> >this. I'm good enough at it, but this isn't what I'm looking to do. I
have
> >higher priorities. Microsoft will write that program, I'm sure.
>
> Actually I don't. I worked as part of a team for a startup company
> that is no more. I created and scripted for the agents. Others wrote
> wrappers around the control, chat clients and servers, and so on. We
> were limited by the direction given to us by our boss. We could have
> accomplished a lot more had the company focused more intently on MS
> Agent for MS Agent's sake instead of their lofty goals of interactive
> television, but they chose to look at MS Agent as a stopgap substitute
> for future technology. Consequently, they missed out on some
> opportunities to be clear leaders in MS Agent development, more
> specifically with their chat client/server.
>
> However, the experience was enough to show me what the potential was.
> I am in no personal position to rewrite what was written at I Pyxidis,
> nor to buy out their I.P. myself.
Obviously you aren't prohibited from adding input to further projects. Why
not open up new doors of opportunity instead of trying to slam them shut?
> I would like to create some more characters and make them freely
> distributable, and perhaps find either a vehicle to purchase I
> Pyxidis' I.P. or secure funding to write a new chat server, but that's
> dependent on whether my dayjob and my professional connections
> provides such an opportunity. I certainly don't have the time nor
> inclination to pursue MS Agent development fulltime outside of my
> dayjob.
Ok. Believe me, I can understand the obstacles software developers face more
than anyone. I would suggest perhaps using some of your time to write
Lightwave tutorials for making Agent characters and put them on an
affordable pay site. I know I'd spend some hard earned money to learn in
that direction, I'm sure you know more about it there than anyone else. It's
said that 'Those who can't, teach.' I think you've kind of reached a point
where you can't afford to progress on these sorts of things, and don't have
the time to do so, perhaps you could find a way to use what you've learned
to benefit others. If I reach a point where I am at a stalemate and can't
move forward, I will do the same.
I have a lot of great things in mind for Agent and they will all be free or
extremely affordable. I want these things to make it to market and I
strongly believe this technology has the capability to change the world.
That's why I've made it my mission to learn everything about it, so I could
find a way to use it in a way that will create something self sustaining so
I can develop free resources for those who will benefit from it.
> >Use another technology
>
> There is no other technology. There could be, but there isn't. And
> that's what's the most frustrating. You'd have to license some tech
> here, some tech there, and glue it together as something new, and I'm
> in no position to do that. I've researched plenty including just
> about every 3D web technology out there. None of them have the right
> combination of features to provide a substitute for MS Agent out of
> the box. And there was no way to glue technologies together together
> to hack it to work, either.
I really don't know what to tell you here. You're suggesting MS change
everything and thwart the development efforts others have put together. I
don't necessarily agree with that.
I just took a minute to walk outside and catch my breath. I'm really hot to
deliever what I've been working on, and I don't want you to catch the brunt
of my agression. I come into this group, and see the messages from
Dickerson, who only wants to complain about MS and tries to destroy the
technology, and I get really bent. And then I see you suggesting a lot of
what I've put together and what other people have created is already
obsolete, and my 'take no prisoners' mode kicks in. It's not personal.
> I wish some entrepreneur would step in and fill that void, but so far
> it hasn't happened. There is still this wide gap between developers
> who write fancy 3D plugins for web browsers and what those things can
> do, and the kinds of awesome graphics you see in PC games. If you can
> merge the two, you've really got something. Plenty of PC games
> network through TCP/IP. Some games like Everquest et. al. are only
> playable with an internet connection. So the lines are blurred. It
> must be that nobody wants to just give away their technology as a
> general purpose 3D construction set. They want to keep it
> proprietary. So we're left with cumbersome beasts like 3DML.
> >there, I will expand into more complex themes which require 'innovation'.
> I can't wait.
For me, it's just a matter of resources. I have the motivation and alot of
the know how. But I'm hesitant to release this little piece or that idea
until I know it works well enough to go, "Yeah, that's a great thing, I
think people will use it and like it."
> I still use an Amiga system.
I loved the Commodore way of doing things. They were truelly ahead of their
time.
> Have you ever played with one of the Catz or Dogz games? Very similar
> to MS Agent. I've actually met the author. Those things rely on a
> foundational graphic engine, the ballz engine. Can't you see the
> overlap here? It's not apples and oranges.
> You are the one taking this personally, and I can see why. I do
> respect you, but I just disagree with your resistance against newer,
> different types of agent technology, and the imporance you place on
> the "barely computer literate with old PC running 16 colors"
> demographic as Agent's target demographic.
I can't really define Agent's target demographic, that's Microsoft's job,
but they seem to always try to get the least common denominator in there
somehow. I really think Agent is here so Grandma can more easily understand
her computer, so John can more easily book his appointments, and so Roy can
make his presentation with less effort. If it's here for games or for 3D
this or that, Microsoft will prove me wrong.
> Your dedication bests me, I'll give you that. I just hope for your
> sake it pays off. MS Agent has been a glorified hobby for me, not
> intended to be my bread and butter. I'd guess that is also the case
> for most MS Agent developers.
Perhaps I enjoy my glorified hobby too much. As for it paying off, I think
it has in that it's given me something to do with my time while the majority
of other folks I know were out wasting their lives watching TV. I've learned
software design and a great deal about people in the process of what I've
been trying to do here. If I never get a job writing software, and if I wind
up working at 7-11, I'll have more money than I've had in the last 5 years,
and I'll have lived my life knowing that I at least tried to accomplish
something.
I started out on this road homeless, with only a few skills but a lot of
understanding of computers, and I've made it as far as across the street
from Microsoft. If I make it no farther than that, I've at least had a lot
of fun, lived just a little more than the people I've known, and been just a
small part of bringing the future closer. If I make it another week, I will
consider myself a success, and that is more important than any of this ever
has been.
I've been the protagonist in my own grand adventure, and in the process, the
people I've met, the help and insight they've given me, have made me more
rich than Bill Gates could ever possibly be. If I ever make a cent, they'll
get more than half of it, and the reason why I am so gung-ho is because I
want to repay them for their kindness, I want to share my life with them, no
matter where I am, and this is the only way I can see that I might be able
to do it.
> >That's why this trip is important to me, and why I've told everyone here
>
> What's this trip all about?
This trip is about my desire to see Microsoft, my goal to expand on a
technology for the benefit of people around the world, and a test, to see if
it is true that the freedom to innovate exists.
It's the end of an enormous circle in my life that started the first day I
ever sat down in front of a computer, and the beginning of the rest of it.
> >End Of Line.
> Is that a Tron reference?
:)
Glenn Saunders
<thegre...@hotmail.com> wrote:
>I just took a minute to walk outside and catch my breath. I'm really hot to
Frankly, it's not normal to get this bent out of shape in a newsgroup
debate. Anyone who is a veteran on the internet needs to know how to
moderate their passions in flamewars. But anyway, my comments alone
aren't going to threaten the future of MS Agent or your projects. I'm
merely voicing an opinion.
You've clearly got a lot of yourself wrapped up in MS Agent 2.0, and I
can respect that. This thread really doesn't have much life left in
it so I'll let you have the final word and move on.
Rob Lindman
> On Mon, 1 May 2000 16:17:33 -0700, "Rob Lindman"
> <thegre...@hotmail.com> wrote:
> >I just took a minute to walk outside and catch my breath. I'm really hot
to
>
> debate.
Correct. Things stopped being normal for me the day I started talking to my
computer, and a little copy of my own head started talking back.
I've taken a laxative, now I can get back to work.
Rob
elbertdee
talking about.
I created a character and he does what I want if I give him the
instructions. Since I had some specialized animations beyond the standard
set I added them to the character. I needed fire so I created one , it has a
spark animation.
If I want my character to walk then I make the animation so when he moves it
looks like its walking.
I am sorry, if Ive misunderstood the discussion but for my purposes it has
allowed me ti finally put an application together in the manner Ive been
trying to figure out for several years.
My characters talk walk and obey command, voice too so ...
Ill shut up now
Elbert dee Walston
--
-----------------------------------------------------
Click here for Free Video!!
http://www.gohip.com/free_video/
Glenn Saunders <cybp...@earthlink.net> wrote in message
news:390d6410...@news.earthlink.net...
James Driscoll Jr
"Rob Lindman" <thegre...@hotmail.com> wrote in message news:#3zK1yBt$GA.299@cppssbbsa04...
Microsoft Agent
James Driscoll Jr <spac...@resistanceisfutile.com> wrote in message news:u0cGJCFu$GA....@cppssbbsa02.microsoft.com...
Ed Ross
>You find that the SR consumes a significant amount of processing horsepower
in listen constantly mode.
Did you try on other another OS, not just Windows? Perhaps another OS may
not take up processing power that could be given to the SR system? Take a
look at http://www.theregister.co.uk/000505-000008.html for an example. If
2GHZ is needed for the OS, what hope is there for extras?
Thanks,
Ed Ross
The Agent LaunchPad -- http://www.agent-launchpad.co.uk
MS Agent VB tutorials.
AgentVision -- http://www.agentvision.co.uk
The AgentVision Show, downloads, polls and more!
James Driscoll Jr
"Microsoft Agent" <msag...@microsoft.com> wrote in message news:#z20rzIu$GA....@cppssbbsa02.microsoft.com...
Microsoft Agent
James Driscoll Jr <spac...@resistanceisfutile.com> wrote in message news:e20#uwOu$GA.313@cppssbbsa04...