Loading ...
Sorry, an error occurred while loading the content.

Re: [hackers-il] Mouse vs. keyboard

Expand Messages
  • Nadav Har'El
    ... I m sorry, but I completely don t buy this research (but note that I haven t read it - I m just speaking theoretically ). Are you sure this wasn t one of
    Message 1 of 7 , Oct 27, 2000
    • 0 Attachment
      On Thu, Oct 26, 2000, Chen Shapira wrote about "[hackers-il] Mouse vs. keyboard":
      >...
      > Each and every user was able to perform the task using the mouse
      > significantly faster, an average of 50% faster.
      >...

      I'm sorry, but I completely don't buy this research (but note that I haven't
      read it - I'm just speaking "theoretically").

      Are you sure this wasn't one of those "Scientific Maariv" articles? :)


      The efficiency of using the keyboard vs. the mouse depends on the features
      the OS offers the user, so it doesn't make sense to talk about this issue
      without comparing actual operating systems.

      For example, to copy a file on a Windows machine, it is obvious that
      drag-and-drop is quicker than opening a DOS window and using the "CD", "DIR",
      and "COPY" commands. However, how do you, using the "faster" Windows Explorer,
      copy all the GIF files under some directory (and its subdirectories) to
      another drive? You'll probably be forced to do it one by one, and believe me,
      while using the mouse is very efficient for doing one thing, it because
      horribly inefficent when repeating an action several dozen times. On the
      other hand, in Unix, using zsh (for example), I just do
      cp /some/directory/**/*.gif /another/directory
      which is much faster than the mouse way.

      Another example, suppose you compare the ease of printing a Word document
      (click the load icon, choose file with mouse, click print button) to
      the difficulty of printing a tex file (suppose for a moment that you don't
      have a GUI for tex) - typing "latex" and "dvips" commands. Is this really
      a good comparison? No. Why? First, people tend to repeat commands (you print
      the file, pick up the printout, find an error, correct it - and then you
      print it again), and modern shells make it very easy to repeat commands.
      Second, what happens when you need to print, say, 10 documents? It takes
      me less than 30 seconds to type
      for i in *.tex
      do
      j=`basename $i .tex`
      latex $j
      dvips $j
      done
      and it will take me much more than that to load and print 10 files into
      Word (forgetting that you can probably do that in VB - VB is more commonly
      used by virus writers than actual users).

      People may argue against what I said by saying that typical users don't
      need such sophisticated loops and repetitions of commands. I say they're
      wrong - beginner, unsophisticated, users may not need them, and find Windows
      very convenient. But once you learn a little more about your computer, Windows
      starts being a strait-jacket. Some users learn to accept that strait-jacket,
      thinking it is natural, but others (like me) refuse to accept it.

      I once spent half a day trying to insert about a 100 pictures (each about 1 MB
      in size) to a report I was writing in MS-Word. It was hell. I had to manually
      resize, load, manipulate, etc., each and every picture, most of which were
      nearly identical. After several hours, I gave up. Instead, I printed an empty
      page (except the headers) to Postscript, and then wrote a shell script (took
      me about 30 minutes) which ran various commands on each image, transformed it
      to Postscript, and overlayed 3 pictures on each blank page (after changing the
      page number). The result looked very good, and it also allowed me to switch
      the entire set of 100 images at a later time, without any more work. Now that
      is what I call efficiency. It wasn't easy, but it was quick (30 minutes
      vs. hours of work), even quicker to do again if you find an error (just run
      the shell script again), and not boring (for some reason the article you
      quoted thinks that "engaging the mind" is a bad thing. I disagree, and find
      repetitive, boring, tasks much worse).

      In short my theory is that keyboard (command line) UI lends itself much
      better to repetition and autmation, which makes it more efficient in the
      long-run than mouse (graphical) UI for experienced users. Of course, having
      both kinds of interfaces implemented well (see, for example, Emacs) is the
      best thing, because it wins in both contests: it allows for easy one-time
      tasks, and easy automation.


      --
      Nadav Har'El | Friday, Oct 27 2000, 28 Tishri 5761
      nyh@... |-----------------------------------------
      Phone: +972-53-245868, ICQ 13349191 |War doesn't determine who's right but
      http://nadav.harel.org.il |who's left.
    • Adi Stav
      ... Interesting. Do you have a URL for that? - Adi Stav
      Message 2 of 7 , Oct 28, 2000
      • 0 Attachment
        On Thu, Oct 26, 2000 at 06:15:09PM +0200, Chen Shapira wrote:
        > To all hard-core hackers claiming that using the keyboard is much faster
        > than using the mouse - here's what a study (conducted by apple, one must
        > say) reports:
        >
        > In one study of this phenomenon (Tognazzini, Tog on Interface, 1992.), users
        > were asked to do the same task using the keyboard and the mouse. The
        > keyboard was powerfully engaging, in the manner of many videogames,
        > requiring the user to make many small decisions. The mouse version of the
        > task was far less engaging, requiring no decisions and only low-level
        > cognitive engagement.
        >
        > Each and every user was able to perform the task using the mouse
        > significantly faster, an average of 50% faster.
        >
        > Interestingly, each and every user reported that they did the task much
        > faster using the keyboard, exactly contrary to the objective evidence of the
        > stopwatch.

        Interesting. Do you have a URL for that?


        - Adi Stav
      • Chen Shapira
        ... The reference I gave was to one of the more respected books in the HCI field. But as I said, the research was conducted for Apple. ... Not true. One can
        Message 3 of 7 , Oct 29, 2000
        • 0 Attachment
          > Are you sure this wasn't one of those "Scientific Maariv" articles? :)
          >

          The reference I gave was to one of the more respected books in the HCI
          field.
          But as I said, the research was conducted for Apple.

          > The efficiency of using the keyboard vs. the mouse depends on
          > the features
          > the OS offers the user, so it doesn't make sense to talk
          > about this issue
          > without comparing actual operating systems.

          Not true. One can give a similar task that can be done in a similar way with
          both a mouse and a keyboard, one can test the speed of working with the
          device, as opposed to the power of the OS.

          There are wide diffrenced between mouse based OS's.
          Grabbing a toolbar is much faster in Mac than under windows, the multi-level
          menus are also faster under the mac.
          I'm talking about a task, not about an OS.

          > For example, to copy a file on a Windows machine, it is obvious that
          > drag-and-drop is quicker than opening a DOS window and using
          > the "CD", "DIR",
          > and "COPY" commands. However, how do you, using the "faster"
          > Windows Explorer,
          > copy all the GIF files under some directory (and its
          > subdirectories) to
          > another drive? You'll probably be forced to do it one by one,
          > and believe me,
          > while using the mouse is very efficient for doing one thing,
          > it because
          > horribly inefficent when repeating an action several dozen
          > times. On the
          > other hand, in Unix, using zsh (for example), I just do
          > cp /some/directory/**/*.gif /another/directory
          > which is much faster than the mouse way.

          Why no, I'd use the Find tool, do find for *.gif in the directory, get all
          the files I need (find is recursive), and the just select-all and copy to
          where ever I need.

          It's about knowing the tools, not many people think of the ** trick (I used
          to write shell scripts to do what you just described)

          About the printing example - again each OS is good in specific things,
          batch prints is easier under Unix (although its unfair, more unix users know
          shell programming than windows users know VB), but printing a specific
          segment from a documents is easier under windows.

          In anycase, those are OS attributes, nothing to do with the subject of the
          test.

          > People may argue against what I said by saying that typical
          > users don't
          > need such sophisticated loops and repetitions of commands. I
          > say they're
          > wrong - beginner, unsophisticated, users may not need them,
          > and find Windows
          > very convenient. But once you learn a little more about your
          > computer, Windows
          > starts being a strait-jacket. Some users learn to accept that
          > strait-jacket,
          > thinking it is natural, but others (like me) refuse to accept it.

          The fact that some people never learn to use windows doesn't make it a
          strait-jacket. Once you learn the OS - its as good and usefull as a Unix.
          I'm talking about NT 4.0 here, because thats the only windows version I've
          used.


          > quoted thinks that "engaging the mind" is a bad thing. I
          > disagree, and find
          > repetitive, boring, tasks much worse).

          I never said that one shouldn't ever ever program, instead he better
          repeatedly use the mouse in stupid ways.
          I only quoted an article saying that:
          1. A task that can be performed with and without a mouse in similar
          ways can be done with a mouse.
          2. You may never know it.

          If postscript does the job faster than Word - go ahead. Some people may do
          it in VB, but again, it isn't related to the mouse VS. keyboard argument.

          Engaging the mind isn't bad, it's just misleading, it makes time fly.
          Remember the time when you were programming, and were very suprized to see
          its morning already? thats what i mean.


          > In short my theory is that keyboard (command line) UI lends
          > itself much
          > better to repetition and autmation, which makes it more
          > efficient in the
          > long-run than mouse (graphical) UI for experienced users. Of
          > course, having
          > both kinds of interfaces implemented well (see, for example,
          > Emacs) is the
          > best thing, because it wins in both contests: it allows for
          > easy one-time
          > tasks, and easy automation.

          The other example is NT with Cygnus :-)
        • Gaal Yahas
          ... In case you haven t read it yet, take a look at Neal Stephenson s essay, In the Beginning was the Comamnd Line. It has been published in book form (Hearst,
          Message 4 of 7 , Nov 1, 2000
          • 0 Attachment
            On Thu, Oct 26, 2000 at 06:15:09PM +0200, Chen Shapira wrote:
            > To all hard-core hackers claiming that using the keyboard is much faster

            In case you haven't read it yet, take a look at Neal Stephenson's
            essay, In the Beginning was the Comamnd Line. It has been published
            in book form (Hearst, 1999), but is also available online at:

            http://www.cryptonomicon.com/beginning.html

            --
            Gaal Yahas <gaal@...>
          • Oleg Goldshmidt
            ... And therefore is suspicious. I don t know much about MacOS [flame away], but it is common knowledge that it is very much mouse-oriented. So it is
            Message 5 of 7 , Nov 7, 2000
            • 0 Attachment
              Chen Shapira <chen@...> writes:

              > But as I said, the research was conducted for Apple.

              And therefore is suspicious.

              I don't know much about MacOS [flame away], but it is common
              knowledge that it is very much mouse-oriented. So it is conceivable
              that mouse operations are quite convenient, and command line isn't.
              And this makes me suspect that Apple might have a vested interest in
              "proving" that using the mouse is generally faster.

              I wonder what kind of difference the completion features of modern
              shells (e.g. bash) make in assessing the efficiency of command line
              work.

              > Why no, I'd use the Find tool, do find for *.gif in the directory, get all
              > the files I need (find is recursive), and the just select-all and copy to
              > where ever I need.

              Bad example. Try to grab files whose names match a regular expression
              and that have been modified since the previous Sunday. I don't add,
              "files that you own and can execute, and only those that are not
              symlinks", because that would definitely be too hard for "find".

              This is [partly] tongue-in-cheek, of course.

              > printing a specific segment from a documents is easier under
              > windows.

              What do you mean by "document"? I guess I see what you mean. This
              seems to me related to the fact that Windows documents are created by
              word processors, while unixoids like myself work with text processors
              (not that I have any significant experience with *roff, but I'll stick
              to (La)?TeX for the time being, thank you).

              In any case, I believe I have extensive experience both in reading and
              in writing documents of various kinds and formats, and I don't recall
              a single case when I needed to print a piece of a document that was
              not an integer number of pages. Printing a page or several pages is
              usually easy enough even under UNIX.

              > > Of course, having both kinds of interfaces implemented well (see,
              > > for example, Emacs) is the best thing, because it wins in both
              > > contests: it allows for easy one-time tasks, and easy automation.

              Hear! Hear! ;-)

              --
              Oleg Goldshmidt <ogoldsh@...>
              "... We work by wit, and not by witchcraft;
              And wit depends on dilatory time." [Shakespeare]
            Your message has been successfully submitted and would be delivered to recipients shortly.