Software Agents in Leopard

Posted:
in macOS edited January 2014
I keep coming back to the 1987 Scully era Knowledge Navigator. A prototype device way ahead of its time. It predated the Internet and hypertext but essentially predicted its use.



It also featured an advanced text to speech system which made the computer sound human. It also featured a speech recognition system that allowed a user to be able to communicate with the device.



The interesting thing about the software and underlying operating system was the bow tie wearing software agent that communicated via speech to the user and the user to the agent. The agent was able to announce a phone call and to take a message.



This is could easily be something in the ichat answering machine feature of leopard. The Knowledge Navigator (Shorthand KN) also had an ichat like camera mounted in the unit.



I would like to throw out to the membership a question. Are we down the technological road far enough for Leopard to have many if not all of the KN features. Are software agents developed enough for mainstream use? Microsoft's BOB was a dismal failure. Could Apple pull it off?



What Super Secret features could Apple pull out of its hat that would make us all stand up and cheer? Would software agents be that OS improvement that would allow Apple to finally eclipse (get it!! ECLIPSE!!) MS? and it's vista debacle.



So, what do you all think? Or, has the Kool aid finally reached INsanE proportions in my bloodstream?



Knowledge navigator here: http://en.wikipedia.org/wiki/Knowledge_Navigator

Comments

  • Reply 1 of 12
    This sounds like a bad idea, regardless of its technological feasibility. Anthropomorphic software solves few actual software problems while piling on new ones.



    When I get home, I open my PowerBook and within seconds I get the following notifications: new mail, feed updates (news, blogs, pictures from contacts, etc.), calendar appointments, software updates. I see my online contacts and should I wish to view any of these data, they're a click away.



    Now compare that to opening my PowerBook to be greeted with the following voice announcement: "Hi, Neven. You have eleven new mail messages. Message one is from Daniel, titled "Movie Friday." Message two is from Christa, titled "Hello." (...) The following feeds have been updated..."



    When software needs to notify me of something, the notification should be both noticeable and subtle. Voice systems are definitely not the latter.



    When I want to engage in a task - web browsing, picture editing and organizing, playing some tunes - it's infinitely simpler for me to click and type for 3-5 seconds than to explain verbally what I'm trying to do. Opening my last iPhoto import and editing the pictures that appear, at a glance, to be too dark, takes a few seconds and a click per picture. Now try verbalizing that action. Another example: checking the spelling of the word "sleuth" takes me 3 seconds. Now how would I "ask" my Mac for it?



    Just some quick thoughts; in short, I would be stunned if Leopard (or, indeed, any major Apple product) followed the KN metaphor.
  • Reply 2 of 12
    tednditedndi Posts: 1,921member
    Upon Waking-"Hi Neven welcome back. You have 11 new mail messages Your mother, your girlfriend and 9 others for Viagra" Your mother and girlfriend are flagged via your ichat important people list.





    Via ical- "Neven, remember to call your sister it's her birthday."



    All before you take a piss after getting in the house.





    Must we retain the paradigm of the typewriter? That is so 19th century.



    I am talking about a 21st century interface.





    Think about this page from an early Macintosh ad









    or this page from the same ad.



  • Reply 3 of 12
    kickahakickaha Posts: 8,760member
    Quote:
    Originally Posted by nevenmrgan View Post


    This sounds like a bad idea, regardless of its technological feasibility. Anthropomorphic software solves few actual software problems while piling on new ones.



    True.



    Quote:

    When I get home, I open my PowerBook and within seconds I get the following notifications: new mail, feed updates (news, blogs, pictures from contacts, etc.), calendar appointments, software updates. I see my online contacts and should I wish to view any of these data, they're a click away.



    Now compare that to opening my PowerBook to be greeted with the following voice announcement: "Hi, Neven. You have eleven new mail messages. Message one is from Daniel, titled "Movie Friday." Message two is from Christa, titled "Hello." (...) The following feeds have been updated..."



    When software needs to notify me of something, the notification should be both noticeable and subtle. Voice systems are definitely not the latter.



    When I want to engage in a task - web browsing, picture editing and organizing, playing some tunes - it's infinitely simpler for me to click and type for 3-5 seconds than to explain verbally what I'm trying to do. Opening my last iPhoto import and editing the pictures that appear, at a glance, to be too dark, takes a few seconds and a click per picture. Now try verbalizing that action.



    Open iPhoto application. Open last imported album. Edit mode. Lighten. Next. Next. Lighten. Next. (etc)



    While I agree it's cumbersome, it's not impossible.



    I completely agree that voice input is generally slower than any other form in current use, and it comes with a whole host of problems and obstacles. But one place where it is highly useful is when you're doing something else. I can be making coffee, and listen to the computer reel off the days schedule, inbox, etc, and respond to simple requests: "Read my schedule" "List new mail" "Summarize first new mail" or heck, "Shut the hell up" The point being that voice control is only really useful if it's for simple things while you're doing something *other* than primarily interacting with the computer.



    Will it make it to market? It already has. Check out Speakable Items in your Mac. It's far from perfect, but the basic functionality is there.



    Will it be a revolutionary new way of interacting with the machine? Well, it hasn't so far... and I don't see that changing.
  • Reply 4 of 12
    tednditedndi Posts: 1,921member
    Thanks Kick.



    Alright then, where are the trends pointing to to get computing into the next paradigm shift.



    The GUI is a construct of PARC thinking in the 1970s and so is the mouse.



    We are now working with basic refinements of that idea and in essence have gone from the typewriter to the dedicated word processor.



    Where is the next step?



    Will Leopard have any of it or continue as refinements to this basic model.



    Remember Apple can be far thinking even if in its implementation it is too quick to market.



    Digital Camera and the newton are but two examples.
  • Reply 5 of 12
    Quote:
    Originally Posted by nevenmrgan View Post


    This sounds like a bad idea, regardless of its technological feasibility. Anthropomorphic software solves few actual software problems while piling on new ones.



    When I get home, I open my PowerBook and within seconds I get the following notifications: new mail, feed updates (news, blogs, pictures from contacts, etc.), calendar appointments, software updates. I see my online contacts and should I wish to view any of these data, they're a click away.



    Now compare that to opening my PowerBook to be greeted with the following voice announcement: "Hi, Neven. You have eleven new mail messages. Message one is from Daniel, titled "Movie Friday." Message two is from Christa, titled "Hello." (...) The following feeds have been updated..."



    When software needs to notify me of something, the notification should be both noticeable and subtle. Voice systems are definitely not the latter.



    When I want to engage in a task - web browsing, picture editing and organizing, playing some tunes - it's infinitely simpler for me to click and type for 3-5 seconds than to explain verbally what I'm trying to do. Opening my last iPhoto import and editing the pictures that appear, at a glance, to be too dark, takes a few seconds and a click per picture. Now try verbalizing that action. Another example: checking the spelling of the word "sleuth" takes me 3 seconds. Now how would I "ask" my Mac for it?



    Just some quick thoughts; in short, I would be stunned if Leopard (or, indeed, any major Apple product) followed the KN metaphor.



    Remember, while they were using the KN as the input device to cause events to happen, I think it ought to be used in a way that makes in unobtrusive. Given the following scenario:



    I turn on my Mac, and I usually walk away. The Mac does all of its opening tasks, and just waits there, waiting for me to return. Once I get back, with my Mountain Dew in hand, he tells me that I have 6 new messages, and 94 messages were sent to Junk. Also, I have set a conference call to occur at 9:30 AM. From there, I just start working.



    I remember that I have a Baseball game tomorrow evening, so I ask my Mac to tell me tomorrow's weather. Now, I haven't stopped working on my normal daily tasks, and he tells me that the high will be 83° F, and partly cloudy. As I continue to work, my wife IMs me, and my Mac tells me that she IMed me. Not a bouncing dock icon, the Mac told me this.



    So I continue working, then my Mac tells me that there is a critical bug found, sent via email. It is marked as a High importance, and that is why he notified me of this.



    So you can see, there is a lot of advantage to having a computer that can understand your verbal commands, and to be able to talk to it in a normal voice, using your native tongue makes more sense so you are more productive.
  • Reply 6 of 12
    tednditedndi Posts: 1,921member
    That's interesting. Would it make any sense to have airport express hubs with microphones for voice interaction with your base computer.



    So, for example you are going to sleep. You tell your computer to set the alarm for 6:45 then put it into background mode. You then tell your computer to play the sleep playlist for 30 minutes then volume down.



    Your computer sits in the living room (or itv) and you have an airport hub with speakers for itunes in your bedroom. You computer could then be more of an extended part of your life rather than having you stuck at your desk to interact with it.



    for $120 each you could put an airport express in every room for added functionality.

    In your living room the itv would be used.



    In a noisy environment, you would sit at your desk and type.



    Or you could use the iphone as the computer input/speech device.



    Pick it up press a button and your base computer is right there for you.
  • Reply 7 of 12
    The one issue that I see as a major factor in all of this is if your household has multiple computers. For instance:



    My wife, she has an iBook. I have an iMac. I would like to tell my iMac that I need to wake up at 7:00 AM (using the whole Airport Express theory that TednDi brought up). How will the computer that I want to respond (my iMac) know that they were the respondent.



    The only solution that I can come up with is if the Computer has a name too. That would be a little odd at first, but I can definitely see the potential.
  • Reply 8 of 12
    kickahakickaha Posts: 8,760member
    Again, already in Speakable Items.



    You can assign a 'key phrase' that, unless it is uttered, the computer ignores everything you say. For a while, back when I was playing with this, I used 'Jeeves'.



    "Jeeves, open iPhoto." etc



    If you had, say, Jeeves and Jester, you could prepend the commands with the name - but then you have to route audio to all computers simultaneously and let them battle it out, etc, etc, etc. Meh.
  • Reply 9 of 12
    tednditedndi Posts: 1,921member
    You forget Kick that not everyone has the technical skills that you do. I am talking about something really apple simple. Perhaps using more natural speech recognition.



    Most people have never been to that menu in the system prefs panel.



    What about a more 3d interface with objects stretching into the screen for depth?



    What I guess I am trying to get at is where we are all going with the UI. not in terms of cosmetics or consistency but in the next evolution of the pc.



    The last could be said to have been the internet and wireless. Where now?



    Where is that next big breakout coming from.
  • Reply 10 of 12
    kickahakickaha Posts: 8,760member
    Oh no, I know exactly what you mean - setting up these systems is a PAIN. I was just pointing out that the *basic* technologies are already in all of our systems, to those who didn't realize it. Hell, most don't realize that a 'tablet Mac' is doable *NOW*. Inkwell is fantastic. We've had it for a couple of years now. All it needs is a touchscreen. (Hmmm... gee, I swear I just read something about that for MWSF... hint: Axiotron.) It's incredible how much tech Apple jams into each copy of MacOS X, and doesn't market actively.



    3D UIs? Nope. Not until we have 3D displays. 3D on 2D has been tried for 20 years, and no one has been able to make it more usable than 2D on 2D. It simply doesn't add anything other than eye candy.



    Next big breakout? Good question. I have my hunches about what it will be for developers, but I have no idea for consumers.
  • Reply 11 of 12
    TednDi, I was thinking about that same thing today. First, we had the Keyboard, then the Mouse in the 80s. The internet was the 90's. So what is this upcoming revolution? I have a real problem thinking it is Web 2.0.



    Also, with the Natural Voice that is a part of Leopard, it would not be that hard of an extension for voice activated commands. While Kickaha is right in what he said, that isn't a new paradigm. The machine has to respond to your commands, just like a normal person might:



    "Mac, send an email to Kickaha about MacWorld 2007, saying:



    Kickaha, I can't wait to see you there. I am so excited about this keynote. Anyways, I'll see you there.



    End Message. Send Message."





    I honestly think voice is the next input device. Also, if you think about it, every Mac now has three things going for it: Ears, to listen to your commands (the Microphone), a Voice (the speakers) to tell you when certain events or actions happen, and now Eyes (the iSight) to be able to recognize you. As long as my Mac doesn't become homicidal, I'll be perfectly OK with my Mac 9000 personal assistant.
  • Reply 12 of 12
    tednditedndi Posts: 1,921member
    From what I understand, Steve's main focus with the mac has been the disappearance of the computer itself. Perhaps that is why he has the love of glass. But also the thin bezels on the Mac Book Pro. The idea that the computer itself fades away. The iMac is a prime example the All in one. Nothing but the screen, keyboard and mouse.



    Now, if we could move away from the keyboard monitor and mouse even for 20% of the time like for listening to music and watching TV/video on our TV's we are essentially moving away from the computer while the computer follows us around. It removes the shackle of having to remain at your desk.



    I think that is what Steve is getting at. Making the computer not so much as a pocket knife but an all purpose tool for whatever you might want.



    Perhaps in this vein that is why the apple starpage looks like the obelisk in 2001.



    A HAL like interface. We have isight and a microphone.



    ...and also the Apple logo postitioned like the obelisk.



    Kind of looks like an iPod, or a closed black iBook no?



Sign In or Register to comment.