Ubuntu should not copy the Mac

Ubuntu should not copy the Mac

Summary: It is past time for open source to become truly innovative. The Macintosh interface is a nice point-and-click interface, but that's all it is. We need something completely different. How about a high quality voice interface, based on Perlbox?


Perlbox main screenshot from Perlbox.orgUbuntu is putting serious investment into improving the interface of its Linux.

Mark Shuttleworth wants Linux to become comparable to the Apple Macintosh, quoting the watchwords of Web 2.0:

  1. Make your site visually appealing,
  2. Do something different and do it very, very well,
  3. Call users to action and give them an immediate, rewarding experience.

Good idea. But the Ubuntu Developer Summit is taking the wrong approach.

It is past time for open source to become truly innovative. The Macintosh interface is a nice point-and-click interface, but that's all it is.

We need something completely different. How about a high quality voice interface, based on Perlbox? (The graphic above is from Perlbox.org.)

Perlbox already has a KDE interface, and three years of work behind it. It may not be all it can be. What could it be with a few million development dollars?

My friend Lamont Wood has been working with voice recognition technology and says some of it is now ready for prime time. By that he means it can be 95% accurate at 120 words per minute.

True, he was testing a proprietary product, but I am first contemplating an open source voice interface, not a voice-based word processor. The parts to do something ground-breaking appear to be here.

Besides, there are millions of visually-impaired folks, like my mom and my friend Jim Pettigrew, who have been totally left out of the computer revolution until now. Why not bring them in?

If Ubuntu is ready to be truly competitive, then that says to me it's ready to innovate. And if voice isn't your favored direction, what is?

Topics: Networking, Apple, Hardware, Linux, Open Source, Operating Systems, Software, Telcos

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Ubuntu does not want to be OS X

    I agree ubuntu should not try to copy the Mac, but my own understanding is that copying the Mac is not Mark Shuttleworth's intention. What he wants to copy is Apple's knack for designing intuitive, productive interfaces. The approach Canonical is taking in this regard is very laudable. They are setting up labs where Linux newbies are observed using ubuntu. What pitfalls the users encounter are noted and improvements to the interface are then suggested and tested.

    Canonical has also sought feedback from the ubuntu community via its "Ubuntu Brainstorm" website. Most of what gets posted on the site is garbage, of course, but so many ideas have been thrown up that no doubt some valuable contributions will be made this way. It is refreshing to see a company so interested in hearing what its customers want and need. So much of the recent history of computing has been developers telling us what we SHOULD want.

    When Palm introduced the Pilot, it was not the first PDA to hit the market, by far. But it easily became the most popular, because Palm put a lot of effort into usability testing of the interface. Prior to release, Palm spent many months having users with paper-based organizers race against users of the proposed Palm. If a function was not faster and/or easier to do on the Palm than with a paper organizer, the engineers went back to the drawing board to redesign that feature's interface. Palm's management appreciated that people would not adopt technology if it did not make their lives easier.

    As far as voice recognition, I hope Canonical avoids it like the plague. It is not ready yet, and won't be for another hundred years or so. Ninety-five percent accuracy may sound good, but that's terrible - that's missing every twentieth word. Even 99.9% accuracy is poor, in terms of usability of an interface. Too much of speech recognition necessarily depends on understanding the context of the situation. And it'll be a long time before machines can do that.

    Ubuntu has made great strides, but they still have a little way to go. Too much of the ubuntu interface still requires that users drop to the shell to perform common tasks with obscure and cryptic Unix commands. Text-based (i.e., command-line) interfaces will always be faster and more efficient for power users of an OS. But they are very unfriendly to the average person who just wants to use a computer to get things done. I think Shuttleworth understands this, and I think he's on the right track to building a Linux desktop for the masses.
    • Ubuntu does not want to be OS X

      Agreed, my understanding is they want a simple, intuitive and coherrent interface for Linux.

      They want all applications to use the same UI conventions. Much like Apple and Microsoft have with their design guides - although nobody seems to really stick with the Windows ones, not even Microsoft sometimes...

      They want a standard "open" button, a standard "save" button etc. not a competition to see who can make the most gawdy design for their application, which doesn't meld with anything else out there.
    • What is this Ubuntu interface you speak of?

      This brings into play the whole desktop environment (KDE, XFCE Gnome) vs what customization distros do to differentiate themselves (The Dell/Ubuntu one seems to be MIA on the netbook having missed the launch and not even mentioned outside the US).

      How much of your usability is determined at the KDE/Gnome level? (This is not about using one desktop, I enjoy having the choice)

      My retired parents have no problems understanding KDE based desktops. People at the retirement home that our group helped acquire used equipment to run Linux dont seem to have a problem.
      My father went to his friends recently out of town and said he couldnt believe how much he missed using Amarok (which isnt my favorite) when he was there.

      But in terms of software usability, oh heck yeah, there is work to be done. The question is how to get so many independant programs to look/behave a certain way.

      I maintain that you have to take your top 10 most used pieces of free software and make them each do the top 10 most frequent things they are used for and do them well and easily in as little steps as possible.
      Garageband and the whole suite of Apple software are often limited but they suit the needs of most of the people. The real pros wont use Imovie but pay for Final Cut Pro. The serious musician will use specialized expensive sofware not Garageband.

      Example: Audacity is a great program but scary to first time users. Why couldnt the software come with two settings? An easy mode for the most common uses like cutting a clip, recording a child and one click away an advanced look for the people who need it.

      It makes me think of my sister in law who has an Acer One netbook and she loves the 4 coloured square kiosk look of the Linpus Linux. Its fun, cheery, easy to navigate, just what a newbie wants. Her husband uses the XFCE desktop that's underneath and installed Gimp and other software he needs.

      Why is VLC so popular? Because Its a simple interface. Double click on the screen to get full screen, space bar to pause-restart and never have to think about codecs again.

      The most popular uses HAVE to be easy.
      If I'm using an audio program and I want to cut an audio segement and then save it in a ldifferent bitrate at a lower, I should be able to click Save As or similar, then have the file name appear and the option for different format and a button for the specifics about quality and maybe even give an approximation of the size of the saved file. Every time I see it, someone tries to re-invent the feel instead of fine tuning a certain way of doing things.

      Then again, since the distro has often little say in the various programs found in it, the whole process sounds like herding cats.
  • This is time I'll never get back

    If Linux was actually easy to use, what would all the geeks do with themselves. They would lose that sense of superiority and end up pumping gas somewhere...
    • not really

      They will start saying that Linux is bloated. That it is a monolithic kernel and therefore crap. That it includes drivers for every possible hardware even if you don't need it. That ALSA is not good enough for multimedia. They would start complaining that X is older than my grandma and bloated, offering functionality like remote X that no one needs and it doesn't even work with OpenGL anyway. That X is crap because support for video functions that are standard in other OSes since 10 years ago, like "theater mode" when watching videos, are not possible. They will bash that the desktop interface from Ubuntu is based on a toolkit that uses C, something that is prehistoric by now. That there are too many flavours around doing the same thing. Well, and many many more things.

      As a results, they will work on another OS from the ground up. As it becomes more stable they will deploy it on servers everywhere while advertising how their jobs is the most important one in IT industry. They will regularly post in slashdot saying how cool they are because they've did a "make clean" when fired from a company without telling anyone else how to recompile the dark distro from scratch.

      History repeats itself.
      • ;-)

        You must remember that Linux is not Ubuntu.

        I think you've got a point about the elite geeks. But you're "ground up", non-bloated OS already exists, and is called [url=www.linuxfromscratch.org]Linuxfromscratch.[/url]
  • Voice?

    Hey, if the idea is to have a great interface for the visually impaired, maybe we should do Braille as our standard interface. That makes as much sense as voice I/O, never mind Bill Gates' fascination with voice.

    As for me, the <b>last</b> freaking thing I want is a computer that I have to talk to. The office is noisy enough as it is, I <b>like</b> being able to use the computer and the phone at the same time, and if I'm awake in the middle of the night I don't want to wake others up by yammering at the silly puter.
    Yagotta B. Kidding
    • Good point for sure.

      Far too often I have seen the rush to employ a certain type of tech just because its available, or at least can be made to work, with little to no regard to the actual real life on the ground practical application of that tech.

      Voice actuated computing is all very nice, but we still live in a world where most of the time that wouldn't be practical, workable or even wanted.
      • RE: Good point for sure

        Yes, good point. I used a Mac Quadra 840av back in 1993, and it could be controlled via voice. Although not perfect it did work and worked reasonably well as long as it was very, very quiet. Any background noise like somebody else talking, it would get confused and either do something totally different or nothing at all. Nice novelty, but no practicality, especially in a busy office or a busy household with noisy kids.
        • There once was an IBM demo...

          Many years ago, IBM was going to be demonstrating Dragon on a PC to a large crowd, and while the crowd was sitting down, some joker in the front row yelled "Format C:". Another yelled "yes" (All data on drive C: will be lost, are you sure? Y/N) and the software worked as advertised, thus putting a quick end to the demo...
    • I agree.

      I don't want to talk to my PC. I love both GUI's and the command-line and switch around depending on the task.

      Also 95% accuracy is pretty lousy. That means 1/20 times it is wrong.

      Really the main thing is to focus on being intuitive and helpful. Make it easy to organize documents and photos and retrieve them.

      Now for those who have various impairments things like voice activation are a tremendous help. But not for most of us.
    • Vista has decent voice recognition. There are many applications.

      Hands free operation of electronics in the car. Microsoft Sync is nice. <br><br>
      In healthcare, if voice recognition can get to the point of handling all dictation, it would save many millions of dollars in healthcare costs.
      Some physicians already use certain packages fairly successfully. <br><br>
      I'm not sure the low din of low voices into a headset is any "noisier" than the clacking of keyboards actually. I don't think it would be any more distracting. <br><br>
      Eventually with the computer becoming more a part of the home, there could be many applications, and i'm sure clever people will find many more. <br><br>
      Finally, i know it's beyond you to hide your obsessions, but i would stake all i have that Bill Gates is not the only person in technology, including OSS, Apple, SUN, Google that is "facinated" with voice.
      • Speaking of "not hiding your obsessions", Windoze Bigot xuniL_z....

        Is this "projection", like your hero John McCain claiming Obama is running a "slimy" campaign when it's McCain who's slime all the way, do you suppose...?
        • What the hell is Windoze again?

          Who cares about "slimy" American politicians. They are all the same.
          • Exactly....

            Boring "slimy" American politicians are all the same. But i guess we can be thankful for that. Just a few reminders of what slimy and wretched and ruthless politicians can do. <br><br>
            Imperialistic British politicians out to conquer the world by force with her navy<br><br>
            The "crusades" <br><br>
            World War I <br><br>
            World War II <br><br>
            pillage and rape<br><br>
            Just to name a few. <br><br>
            Yep, they can remain the boring "slimy" politicians forever.
        • Wow.

          How do you know my politics? <br><br>
          You're one of those guys who thinks any Windows user must be a very far right gun toting, bible carrying, war mongering, neo-con. <br><br>
          So if that is true, i guess that makes you a far left, tree hugging, atheist, gay pride sort of fellow. <br><br>
          ummm, give me my gun and bible any day.
      • Voice Recognition in Health Care is Not Ready

        As a physician myself, I have been searching for an at-least-barely-passable voice dictation system for 15 years now. I have spent many thousands of dollars on systems that were sold specifically for the purpose of medical dictation. Everything I have seen has been completely worthless, regardless of the cost.The systems seem to demonstrate well, but when you try to actually use them, you quickly realize it's a lot faster just to write everything out longhand.

        It will be many decades before computers will be able to transcribe the human voice accurately. This video is emblematic of my own experiences with voice dictation (and it was done using Vista's voice recognition engine):


      I was going to write the very same thing.

      The LAST thing on earth I could possibly want is a computer I have to talk to!

      And it's not only about the noise, talking is SLOW.
      In my opinion, Voice-based interfaces only sound appealing to slow-minded people.
    • The point is not to reinvent the wheel

      I thought open source was about innovation. Voice
      commands are, if done right, a real innovation. Of
      course they are not appropriate in all circumstances.
      But neither are mice.
      • What wheel ?

        The reason wheels are round is that they do the job they are supposed to do in the best way possible. Trying to re-invent them is pointless. Innovation just to make something different as opposed to making it more useful is just as pointless. Yes, mice are not always the best user interface. That's why thee are keyboards, and trackballs, and touch screens, and yes even voice (I'm sure I left out more than a few). And no, adding voice is not a real innovation, its been around for years mainly in niches where it works well.
        Hemlock Stones