Re: Is symbolic thinking the silver bullit for programming?


Bruce Lewis wrote:
> An untrained eye looking at a series of Chinese characters will see a
> lot of meaningless lines. Even obvious ones like "big" are only obvious
> after someone has told you what they mean. If I understand correctly,
> someone with a written vocabulary of 2000 characters is not likely to
> guess the meaning of the 2001st one, as they would if it were truly a
> graphical system.

Well, that statement, like so many abou thuman capabilities turns out
to be both true and false. Many moons ago I became slightly proficient
in reading Japanese. I was able to be functionally literate well above
my *actual* literacy level because the Chinese characters *are*
composed from a smaller set of radicals which do have consistent
meanings. Having said that, I was still basically illiterate, having
only reached the 400-character level. The set of core radicals is
large, and frequently common radicals are composed with one-off
ideographs which can throw you way off.

'Not likely to *guess* the meaning of the 2001st?' I think it depends
on your definition of 'likely'. It's likely enough that the writing
system was never truly scrapped (even counting Mao's reforms because -
I am told -the old forms were still used in academia). OTOH, you will
find no-one who will contend with the notion that the Chinese system of
writing is the most difficult in the world.

Gripping hand is that even Chinese writing is still a serial knowledge
encoding - just the individual memes are more densely encoded. Rather
like using packed binary integers instead of BCD or even ascii
representations. So the Chinese character argument has little weight in
the 'graphical programming' debate.

Programming is still primarily a linguistic exercise. Kinetic/graphic
programming is more like some of the 'training' that is done with
robots used in assembly line processes. K/G programming will forever be
stuck to very limited domains.

david rush
--
kinetic frenetic

.

Relevant Pages

  • Re: TextOut() to a DialogBox ???
    ... code size as if it had meaning, ... meaningless as a metric of reuse, productivity, or ... Minimizing the public interface might be a good metric. ... In programming, you can code for size or speed. ...
    (microsoft.public.vc.mfc)
  • Re: To Richard Heathfield: enoughs enough
    ... Hungarian notation, the identifiers are easier than non-Hungarian ... the SAME conscientiousness that uses Hungarian notation in the ... neither subject in any way to management whims and meaning ... there are in programming different discourse communities. ...
    (comp.programming)
  • Re: TextOut() to a DialogBox ???
    ... He actually said code size meaning source code size, ... In programming, you can code for size or speed. ... process to review ONLY the "interface" files (in the true ... Part of the role of a good compiler is to peek across ...
    (microsoft.public.vc.mfc)
  • Internalising new linguistic symbols and patterns
    ... Obscure words with precise meaning ... Something similar happens when programming, ... your program seem unclear but can provide the reader with an insight ... The powerfully general idiom seems ...
    (sci.lang)
  • Understanding descriptions
    ... State class descriptions are a neglected _part_ of AI development ... (including reinforcement learning type approaches to A(G)I ... such ideas to include in programming to describe components. ... class of positions that is meaning thousands or millions of positions, ...
    (comp.ai.philosophy)