Is Python the Esperanto of programming languages?

Isaac To kkto at csis.hku.hk
Sat Mar 22 07:41:06 CET 2003


>>>>> "Peter" == Peter Hansen <peter at engcorp.com> writes:

    Peter> I don't know if "stupid" is exactly the word you should use here,
    Peter> unless you're trying to incite antagonism.  Perhaps you meant
    Peter> "unnecessary"?

No, I didn't mean "unnecessary".  But you may substitute the "stupid" word
in my last message by anything that express my feeling of "highly
dissatisfied---even after learning it very hard I still don't want to use
it, and even after trying to get used it in a particularly simple form
(English) for 25 years now I still don't regularly get them right, and I
still keeps wondering why it is there from the beginning" attitude.
Different people use different words to communicate their feeling, and I
don't expect anyone to get "incited for antagonism", especially when I use
exactly the same word against my own mother tongue.

For those who learnt an inflected language from the first year of their
birth, they swallow at an age of 2 the fact that everytime they want to
utter a sentence they need to know the number, time and gender of every noun
and verb they use.  This is not the case for native speakers of
non-inflected languages: they are regularly left unspecified.  Either it is
implied by the context (when you are talking about the weather today, why
you need to keep reminding your listeners that you are in the "present
tense"?  When you are talking about the things happening in World War II, do
you really think the listener needs to be reminded that it is something in
the past in every sentence you speak?), or the speaker just doesn't want to
find it out (e.g., when the speaker want to tell a listener that the speaker
has bought some apple(s) and at least one seems still remaining so the
listener can eat it, he will say something in the line of "there're apples
that you can eat".  But why the speaker has to know there is one or there
are many?)

The contextual information is still there in inflected languages, so the
first part of above makes inflection an unnecessary cruft (and of course,
even more for agreement).  What makes it "something I hate" is the latter
part.  It forces the speaker to consciously reveal irrelevant information.
(You'd say, but the listener might think that it's relevant!  But that's not
the point: if the speaker can expect that, he will speak in a way that the
listener is given the information.)

Given that we are different, why it is better to have non-inflected language
when designing planned languages?  Just because it is easier to learn
something that you do regularly is unneeded in the new language (and in fact
there is no way to do that), than to tell somebody that this is something
completely new and you have to swallow it if you use that language.

    >> BTW, Yes, Chinese is indeed *much more* resistent in noisy
    >> environments (but for a very different primary reason: it's louder).

    Peter> Ah... do you have any evidence of this, or are you just
    Peter> theorizing?

I don't know whether you consider having conducted lessons in both Chinese
and English when people ask questions is of any value as "evidence".  I
consider that might be heavily biased since I'm much more fluent in my
mother tongue for sure.  In that case, no hard evidence.

    Peter> Nevertheless, I believe the main point is simply that Esperanto
    Peter> is provably *much* easier to learn than most natural languages,
    Peter> and easier to at least some extent than probably all of them.

That ignores the frequency factor.  A language that a speaker uses regularly
would clearly be "many times more easy to use" than another language that he
use rarely.  One normally use native languages more than 70% of their time
(for me, much less as I use English as well... expect around 40--60% for
both).  On the other hand, one can expect to use auxillary language for
around 2% of their time.  One would reason that if an auxillary language is
just 1/40 as complicated as a natural language, people won't choose to use
it.  I won't go that far, but it's still true that if it is not
substantially simpler, it won't succeed.

Regards,
Isaac.




More information about the Python-list mailing list