On Oct 17, 2007, at 4:11 PM, Miguel Paraz wrote: > On 10/17/07, Orlando Andico <[EMAIL PROTECTED]> wrote: >> I don't think voice input and output are optimal. >> >> This is because our brains are primarily vision processors. It's >> actually been proven that our visual cortex can remember things that >> are flashed on a screen for 20 milliseconds, which is far less time >> than we can process consciously. >> >> Translation: a visual I/O method will always be faster than a >> voice-based one. > > Plus I think our hand/finger muscles are much stronger than our vocal > cords. Imagine talking all day to your computer ?!
have you guys tried using the speech recognition on os x? i've actually tried it. it works... after a fashion. in a relatively quiet room, say, "open safari" then the computer responds "opening safari" the browser pops up. you can also say, "open reader"... and the browser points you to your google reader. it has its usefulness and limitations and it can be pretty, "high tech sounding"... then there is the part that when you say "open itunes," then boom! iTunes bounces on the dock. then you say "play" and it starts playing music? forget about the machine responding to your command to "stop" or "open your browser" because the onboard mic will just be swallowed by your speaker's music. but you know keyboard clicking is faster for a lot of tasks, even faster than a mouse. Needless to say, i've turned off that feature. i'm pretty sure that some day, speech recognition would be useful say while you're eating breakfast, you can just command your computer like: "Nevermore, download feeds, email, twit podcast and Joan's music folder to Darkside (your iPod's name)". then you rush off for a shower and going out of the house, you pick up your ipod--- which was sync'd and has everything you need for the commute to work. now that's multitasking. imagine having to spend five minutes on your computer in between breakfast and shower to sync up. huge, time saver, speech recognition would be. i think going forward, we will see User Interfaces that are determined by individual applications. no more desktops, no more panels, fewer "k/start menus", if at all. Like an iPhone/iPod Touch UI, makes sense. Every APP, creates its own buttons and human interactions. Much like what Jeff Han has done with his Multitouch UI. Each app determines how BEST to interact with a user. viewing photos would be different from say, browsing through a music collection. Certainly it is different from say, a future MMDA Chief looking over a satellite map of Manila during a storm, in real time and coordinating disaster and civil defense on a virtual map like any warcraft player does. Or for that matter looking over a cluster of computers and managing them through a graphical UI and using your fingers to just stop/start a service for example. Also, one of those things i want to see in a future Desktop OS is a seamless integration with the web. Like for example: i can have picasa or iPhoto, or any photo app then just drag it to a site and every selected photo from my computer gets uploaded to say, flickr or facebook, or multiply or *cringe* friendster. no more "browse, click, ok, then upload" ritual. my first point being... every data that we want to look into or use requires different ways of interacting with it. there is no single dominant way to interact with the machine. Any future OS has to be highly intuitive and intelligent. how do we create a world like this? i think it comes down to various technologies coming together. as an example of what Apple is doing. it has laid the foundation through its Core Services for easy access to audio, images/video, animation and even automator as it's place in the grand scheme of things. that's what Apple is leveraging. in linux, a concentrated effort to make things EASY to develop and access those technologies is somewhat lacking. For the same reason that people "bitch" about linux not being ready for prime time JUST BECAUSE mp3 support isn't part of the default installation of say Ubuntu, for example (when in fact every os deployed doesn't always include MP3 decoding). If it takes a distro to pay for the license for mp3 decoding to be legally included on a distro, why shouldn't they pay for it? same goes for DVD playback. Same goes for every other codec. Yet, from a business point of view--- that's not really a good thing for distros is it? because it's entering the consumer electronics space. they're not after the consumer market. business doesn't care about those silly little things. From our Fan Boy perspective, we want our linux to be the darn best operating system on the planet. What does this mean for Linux? We've got OpenAl/SDL, OpenGl--- as excellent foundations and they all run on linux. the major distros have them. Our current kernel, i think is more than capable to get the ball running. The whole multitouch thing, i believe there is a Do it Yourself forum with open C code on the net. it works on windows and linux and os x. the code uses blobs and ordinary cameras. the cameras are used to determine where your fingers are. Our filesystems--- a combo of xfs, ext3 and maybe one day zfs and nfs and gfs are all out there, are all good. heck, linux have digital media solutions like kino and renderman running on it, so it isn't a question of is linux good enough. There is indeed a HUGE opportunity for any distro to create an Operating System and ultimately a User Interface that is innovative that can blow people's minds. kde 4 and gnome projects and other efforts are underway to help make this a reality. the drivers coming out of ati and nvidia are certainly promising in terms of advancing the linux state of things. imagine, instead of darwin's mach, you have a linux kernel at its heart and on top of that opengl, openal, pdf/postscript, beryl or whatever it takes to get all that eye candy. Apple has certainly built an OS, from a Unix derivative, so there really isn't any case AGAINST any Linux distro not to do the same--- except profit. An operating system, i think is both like a piece of art and the sum of its parts. even the tiniest, seemingly odd part of it like a scripting language can be essential piece that makes whole machine to work great. And right now building a great linux distro to rival and surpass anything out there can be done. it has been for years. BUT and this is a HUGE but: there is no concentrated effort to come up with one. no distro has a grand vision and is trying for one. no single community has a grand vision, trying for one. you have different scattered--- seemingly unrelated projects innovating on their own. But the final piece? the one that assembles the pieces and ships 'em out: it's not there. it's not like the technology isn't here. That's not to say, people aren't trying or not pushing hard enough. i'm saying different communities have different idea of what should be and shouldn't be but there is no one benevolent dictator that gives the final yes/no or what is. it's like the analogy of too many cooks in the kitchen, with scattered brains all over the place. i think, we fanboys and the general public tend to forget is that Linux is a first a kernel and second a distro. That's part of what makes linux powerful. The distros wrap things around the kernel and they turn it into something special. What a Linux distribution should be? My 2nd point being: Linux can be anything you want it to be. ------------------ Cocoy Dayao "People who are really serious about software should make their own hardware." --Alan Kay _________________________________________________ Philippine Linux Users' Group (PLUG) Mailing List [email protected] (#PLUG @ irc.free.net.ph) Read the Guidelines: http://linux.org.ph/lists Searchable Archives: http://archives.free.net.ph
