Cobol - OT: Linguistics [Was: Cobol work?]

Author

OT: Linguistics [Was: Cobol work?]

Oliver Wong

2005-10-17, 6:55 pm

"Judson McClendon" <judmc@sunvaley0.com> wrote in message
news:zqT4f.35459$Lp.18470@bignews5.bellsouth.net...
> And it has been proven that humans do better with letter based languages
> than with pictogram type languages, which were abandoned by everybody but
> Asians long ago.

"Proven" in the formal sense of the word? Probably not, given the
vagueness of "do better", but I'd still be interested to hear/read about any
research done in comparing alphabetic languages versus ideographic languages
and their effects on human cognition.

I don't really "like" ideographic languages; they require far too much
memorization in my opinion. That being said, ideographic texts seem to allow
for extremely fast reading speeds. Consider the above paragraph in English.
It consists of two two sentences (one short and one longer) and 258 English
characters. Read at a relaxed pace, I'd imagine it would take a typical
English reader 5 or 6 seconds.

The same sentence translated to Japanese might be around 130 Japanese
characters, and could probably read in about 2 or 3 seconds by a typical
native Japanese reader, again at a relaxed pace. My Japanese is not even at
the elementary level yet, but I am often amazed at how fast caption flies by
on Japanese television shows. When watching these shows with my Japanese
friends, I often protest "How the hell could anyone possibly read all that?"
causing them all to stare at me in astonishment, for they had all read the
text with no problem at all.

> Some indication of this particular aptitude of the brain for lettered
> languages can be seen in that the following can be fairly easily read:
>
> -----------------------------
> I cdnuolt blveiee taht I cluod aulaclty uesdnatnrd waht I was rdanieg. The
> phaonmneal pweor of the hmuan mnid. Aoccdrnig to rscheearch at Cmabrigde
> Uinervtisy, it deosn't mttaer in waht oredr the ltteers in a wrod are, the
> olny iprmoatnt tihng is taht the frist and lsat ltteer be in the rghit
> pclae. The rset can be a taotl mses and you can sitll raed it wouthit a
> porbelm. Tihs is bcuseae the huamn mnid deos not raed ervey lteter by
> istlef, but the wrod as a wlohe. Amzanig huh? Yaeh and I awlyas thought
> slpeling was ipmorantt !!!! (evne befro slpel ckech)
> -----------------------------
> --

I've seen some varations of this text, but this is one of the most
impressive examples (to me anyway). I was really astonished I could decypher
"uesdnatnrd", though I wonder if "phaonmneal" was a "cheat" in the sense
that the 'h' followed the 'p'. Would it have been as easy to decypher if the
'h' and 'p' were non-adjacent? E.g. "paaehmnnol"?

I suppose a good test to ensure that the above example wasn't "carefully
crafted" is to write a program to take any plain text file and do the
shuffling itself, to really verify if you can read English text with little
difficulty, as long as the "first char/last char" invariant still holds.

- Oliver

Peter Lacey

2005-10-17, 9:55 pm

Oliver Wong wrote:
>
> "Judson McClendon" <judmc@sunvaley0.com> wrote in message
> news:zqT4f.35459$Lp.18470@bignews5.bellsouth.net...
>
> "Proven" in the formal sense of the word? Probably not, given the
> vagueness of "do better", but I'd still be interested to hear/read about any
> research done in comparing alphabetic languages versus ideographic languages
> and their effects on human cognition.
>

I read SOMEHWERE, at SOME TIME, that it was ironic that GUI and icons
had become so popular because the originators (the Xerox group?) had
come to the conclusion that text-based choices were more easily
understood. If I could tell you the source of this I would. As it is,
it ancedotally supports the above assertion.

PL

Richard

2005-10-18, 3:55 am

> I read SOMEHWERE, at SOME TIME, that it was ironic that GUI and icons
> had become so popular because the originators (the Xerox group?) had
> come to the conclusion that text-based choices were more easily
> understood. If I could tell you the source of this I would. As it is,
> it ancedotally supports the above assertion.

Words are much better if you don't recognise the icon. For example
Windows XP completely changed everything requiring relearning (which I
avoid wherever possible) and Vista probably changes them all again.

Personally I really like KDE 3.4 where the 'icons' are actually
previews so a document 'icon' is the first few words of the text of the
documents - hover the mouse and you see a miniture first page preview.

Howard Brazee

2005-10-18, 6:55 pm

On Mon, 17 Oct 2005 21:18:15 GMT, "Oliver Wong" <owong@castortech.com>
wrote:

> The big problem Japanese tend to have is with names (e.g. family names).
>Kanjis usually have multiple pronounciations with different associated
>meanings (sort of like homonyms in English). When these kanjis are used for
>communication, you can usually tell which meaning (and thus which
>pronounciation) to use via context, but when these Kanjis are used for
>names, you can't rely on the sequence of kanjis to have any meaning, so you
>can't use context to figure out what the pronounciation should be.

Names have this problem in quite a few languages, as families move
from one culture to another either in space or in time.

Howard Brazee

2005-10-18, 6:55 pm

On 17 Oct 2005 19:56:13 -0700, "Richard" <riplin@Azonic.co.nz> wrote:

>Words are much better if you don't recognise the icon. For example
>Windows XP completely changed everything requiring relearning (which I
>avoid wherever possible) and Vista probably changes them all again.

I've experienced the same thing. But I suppose icons are better when
you don't recognize the word.

Michael Wojcik

2005-10-18, 6:55 pm

In article <ryU4f.40597$ir4.6791@edtnps90>, "Oliver Wong" <owong@castortech.com> writes:
>
> The big problem Japanese tend to have is with names (e.g. family names).
> Kanjis usually have multiple pronounciations with different associated
> meanings (sort of like homonyms in English). When these kanjis are used for
> communication, you can usually tell which meaning (and thus which
> pronounciation) to use via context, but when these Kanjis are used for
> names, you can't rely on the sequence of kanjis to have any meaning, so you
> can't use context to figure out what the pronounciation should be.

Also, names sometimes use archaic (in the sense of "stricken from the
language by the Occupation government after WWII") pronounciations
for some radicals and kanji, or kanji that are themselves archaic.
(Archaic or rare kanji - ones that aren't part of the standard set of
about 2000 - show up once in a while in ordinary text as well.)

That's why in printed Japanese you occasionally kanji with furigana -
hiragana or katakana (the two syllabaries) written in small type above
or beside the kanji to tell the reader how it's pronounced. Furigana
is also used in texts for children who haven't mastered all their kanji
yet, and I've seen it used to actually replace one word with another.

I think that in that last case a character was speaking to herself and
referred to her father-in-law; the term "father-in-law" was written
using the kanji for the formal word, but it had furigana for "otousan",
literally "father". I believe the reader is supposed to understand that
she *said* "father" but was referring to her father-in-law. The kanji
represented the concept and the furigana its sound-image.

That would be a tough concept to convey economically with a phonemic
writing system, since the point of such systems (pace Derrida) is to
represent speech. That anecdote - though it proves nothing - does
suggest how logographic/ideographic writing systems can have higher
information entropy than phonemic ones.

--
Michael Wojcik michael.wojcik@microfocus.com

Advertising Copy in a Second Language Dept.:
Tapestry of the encounting and the farewell, the birth and the death.
You can hear the human being's song running through the 100 years.
-- Squaresoft

Michael Wojcik

2005-10-18, 6:55 pm

In article <ybU4f.40588$ir4.14395@edtnps90>, "Oliver Wong" <owong@castortech.com> writes:
>
> My Japanese is not even at
> the elementary level yet, but I am often amazed at how fast caption flies by
> on Japanese television shows. When watching these shows with my Japanese
> friends, I often protest "How the hell could anyone possibly read all that?"
> causing them all to stare at me in astonishment, for they had all read the
> text with no problem at all.

I sympathize - my Japanese has never been good enough for me to read
subtitles or film credits, though I occasionally catch something as
they fly by.

This reminded me of a developer I had working for me some years back.
He was fresh out of college with a major in physics and a minor in
Chinese. We were in the habit of watching kung fu movies (in half-
hour or so segments) in a conference room during lunch. Many of the
films we got were subtitled in Chinese (dialogue was typically
Cantonese, and there's a large Mandarin-speaking market for these
movies), and either dubbed into English, or subtitled in English as
well as Chinese (so there were two lines of subtitles running along
the bottom of the frame).

He could sometimes follow the dialogue in both languages simultaneously,
and was highly entertained at just how different they were.

(For that matter, we had one film - "Duel to the Death" - with both
English dubbing and subtitles. From different translations. Gave
the film a whole new dimension.)

--
Michael Wojcik michael.wojcik@microfocus.com

Any average educated person can turn out competent verse. -- W. H. Auden

Howard Brazee

2005-10-18, 6:55 pm

On 18 Oct 2005 13:50:43 GMT, mwojcik@newsguy.com (Michael Wojcik)
wrote:

>He could sometimes follow the dialogue in both languages simultaneously,
>and was highly entertained at just how different they were.

Turn on subtitling in English or closed captioning in English (similar
but not identical text), in an English language DvD and note the
differences between the dialog.

LX-i

2005-10-18, 9:55 pm

Richard wrote:
> Personally I really like KDE 3.4 where the 'icons' are actually
> previews so a document 'icon' is the first few words of the text of the
> documents - hover the mouse and you see a miniture first page preview.

Do you use Konq much for file maintenance? Maybe it's just the way I
learned, but I still end up using the shell for most of that stuff.
Maybe I just need to learn about its features. I had noticed that the
icons were thumbnails (for pictures) or had text in them (for text
files) - even PDFs, HTML, and XML files give a preview. Quite

...

Under Windows, my shortcut to start the command prompt has the tool tip
"When you need to get things done NOW!" :)

--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~
~ / \ / ~ Live from Montgomery, AL! ~
~ / \/ o ~ ~
~ / /\ - | ~ daniel@thebelowdomain ~
~ _____ / \ | ~ http://www.djs-consulting.com ~
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
~ GEEKCODE 3.12 GCS/IT d s-:+ a C++ L++ E--- W++ N++ o? K- w$ ~
~ !O M-- V PS+ PE++ Y? !PGP t+ 5? X+ R* tv b+ DI++ D+ G- e ~
~ h---- r+++ z++++ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~

Richard

2005-10-19, 3:55 am

> Do you use Konq much for file maintenance?

I use Midnight Commander for all development work and working in the
development system.

Michael Wojcik

2005-10-19, 9:55 pm

In article <eb3al1t354thh0qa3msi645fu8dfpv04o1@4ax.com>, Howard Brazee <howard@brazee.net> writes:
> On 18 Oct 2005 13:50:43 GMT, mwojcik@newsguy.com (Michael Wojcik)
> wrote:
>
>
> Turn on subtitling in English or closed captioning in English (similar
> but not identical text), in an English language DvD and note the
> differences between the dialog.

True, but these were much more significant differences than what
you typically get with same-language captioning. There were
differences of tone and at times of meaning. Character names
were sometimes radically different. Entire statements would be
inserted into or omitted from the English dub.

Closed captioning is usually deliberately different - because it
generally includes notation for sound effects and other nonverbal
audio elements, dialog is often somewhat compressed to make up for
it. Also, if the DVD was mastered using a CC transcript from a
TV broadcast, those captions are often written on the fly, often
by volunteers who are well-intentioned but not expert, and often
unfamiliar with the material.

Subtitles are sometimes written from the script rather than from
what was actually recorded, and while a decent subtitle editor
will try to catch differences caused by ad lib dialog, on-scene
changes, and the like, sometimes they miss something.

(With foreign-language DVDs, there's often considerable grumbling
among serious fans over "dubtitles", which are subtitles written
from the dub script, rather than from the original-language
dialog. But that's another issue.)

--
Michael Wojcik michael.wojcik@microfocus.com

Thus, the black lie, issuing from his base throat, becomes a boomerang to
his hand, and he is hoist by his own petard, and finds himself a marked man.
-- attributed to a "small-town newspaper editor in Wisconsin"

Howard Brazee

2005-10-19, 9:55 pm

On 19 Oct 2005 17:18:59 GMT, mwojcik@newsguy.com (Michael Wojcik)
wrote:

>True, but these were much more significant differences than what
>you typically get with same-language captioning. There were
>differences of tone and at times of meaning. Character names
>were sometimes radically different. Entire statements would be
>inserted into or omitted from the English dub.

I remember a post in a movie forum where an Italian noted a pun used
in the movie _Young Frankenstein_. He was curious about the English
language version where the pun was between "werewolf" and "where
wolf". The Italian pun was much better, but not at all related to
the English pun.

Good translators need to go beyond direct translation. In books,
where you have the freedom to change the cadence, this is easily
recognized - in movies, often the only freedom is in changing
syllables to not make the mouth movements too silly.

Poor translators are generally cheaper though.

It would be funny to do a take off of the Star Trek Universal
translator where someone makes a one syllable answer - but we all hear
a much longer reply - or visa versa. But the translators only
translate the hard stuff - words that we all know (the Klingon words
for honor, etc) don't get translated. (Sort of like the western
where the Mexican knows English - but common words such as "mister",
or "yes").

Hmmm. How about a universal translator that makes a library based OO
language look like structured CoBOL when I need to debug an
application?

2005-10-19, 9:55 pm

In article <h71dl116up2dji42la76tipbgk5e8ptijo@4ax.com>,
Howard Brazee <howard@brazee.net> wrote:

[snip]

>Good translators need to go beyond direct translation. In books,
>where you have the freedom to change the cadence, this is easily
>recognized - in movies, often the only freedom is in changing
>syllables to not make the mouth movements too silly.

'You insult me, you insult the Shao Lin temple... now we will fight and
see whose Kung Fu is stronger!'

>
>Poor translators are generally cheaper though.
>
>It would be funny to do a take off of the Star Trek Universal
>translator where someone makes a one syllable answer - but we all hear
>a much longer reply - or visa versa.

Something like Chaplin did in 'The Great Dictator'?

DD

Oliver Wong

2005-10-19, 9:55 pm

"Howard Brazee" <howard@brazee.net> wrote in message
news:h71dl116up2dji42la76tipbgk5e8ptijo@
4ax.com...
>
> Good translators need to go beyond direct translation. In books,
> where you have the freedom to change the cadence, this is easily
> recognized - in movies, often the only freedom is in changing
> syllables to not make the mouth movements too silly.

I was participating in a fansub (subtitles written by fans of a movie,
rather than by some other, more "official", entity) project when we had a
discussion about this. The movie was "Ghost in the Shell 2: Innocence", and
we were writing the English subtitles. We had a discussion (it was too tame
to be called an "argument") about whether we should try to replicate every
nuance present in the original Japanese dialog, or if we should try to make
the English subtitles "flow well".

In the end, we decided that our target audience were the "Otakus"
(hardcore fans) who would be more interested in a "purist" rendition of the
story, and we ended up putting a few "translator's notes" sprinkled about
during those scenes when no dialog was taking place.
[color=darkred]
> On 19 Oct 2005 17:18:59 GMT, mwojcik@newsguy.com (Michael Wojcik)
> wrote:
>

I'm reminded of a Japanese movie in which there were some references to
gr

mythology. The subtitler was familiar with Gr

mythology (or did
research), and correctly translated a name to it's standard English spelling
of "Triptolemus". However, unless you knew Triptolemus was the name of a
character from Gr

mythology, due to the way it was pronounced by the
Japanese actor, you'd think they were arbitrarily renaming people.

There also exists some concepts that are difficult to translate from one
language to another (like your "father" Furigana example in another branch
of this thread). In the Ghost in the Shell project mentioned above, we had
to translate a scene were specific reference to honorifics were made
(paraphrased from memory):

Lin: Ba...Batou!
Batou: You say my name without -san or -sama?
Lin: I'm sorry, I didn't do it, Batou-san!
Batou: I'm too good to be addressed merely as "-san" by you!

(My possibly incorrect understanding is that) the -san honorific is
generally used to address someone who is your equal, while the -sama
honorific is usually reserves for royalty or someone you respect very
highly. I can't see how this scene could have been subtitled without one of
the following:

a) Assuming the reader/viewer understands the Japanese honorific system.
b) Change the entire meaning of that scene.
c) Add a "translator's note" explaining the Japanese honorific system
during the opening credits of the movie.

We went with c, again because of our assumptions about our target
audience. I think typically though, most "official" subtitles don't want to
target only the hardcore-fan segment of the market, and so typically choose
b. The idea is that people who need subtitles probably won't know enough
Japanese (or whatever the original language was) to realize that the story
is being changed.

- Oliver

Howard Brazee

2005-10-19, 9:55 pm

On Wed, 19 Oct 2005 18:39:01 GMT, "Oliver Wong" <owong@castortech.com>
wrote:

>Lin: Ba...Batou!
>Batou: You say my name without -san or -sama?
>Lin: I'm sorry, I didn't do it, Batou-san!
>Batou: I'm too good to be addressed merely as "-san" by you!

Without knowing the subject matter, the concept was clear to me.

I've been reading a (very fun) series of SF novels that are basically
Patrick O'Brian novels in space. The author, David Drake puts an
apology/explanation in the forward for using current systems of
measurement. But he also creates dialog where familiarity with the
British class structure is useful. Enough time has happened so that
we know they won't talk like us anyway. I don't see that that
explanation was necessary.

The toughest part about translation is when people think differently.
We see some of that when changing to OO programming. Things don't
translate directly.

We don't convert systems, we re-write systems. The benefit of being
able to recompile our CoBOL in the new system doesn't count for much
when the new system has a completely different data design, and a
completely different work flow.

Michael Wojcik

2005-10-21, 6:55 pm

In article <9pw5f.37137$S4.35399@edtnps84>, "Oliver Wong" <owong@castortech.com> writes:
> "Howard Brazee" <howard@brazee.net> wrote in message
> news:h71dl116up2dji42la76tipbgk5e8ptijo@
4ax.com...

Agreed, though there are some interesting cases. In anime, for
example, dialogue is often recorded after the animation is done, so
there's no attempt to make the mouth "flaps" (as they're called)
agree with the actual spoken syllables. When anime gets dubbed for
US viewers, however, the translators often *do* try to get the flaps
to agree with the audio, so the dub track may match the video better
than the original does.
[color=darkred]
> In the end, we decided that our target audience were the "Otakus"
> (hardcore fans) who would be more interested in a "purist" rendition of the
> story, and we ended up putting a few "translator's notes" sprinkled about
> during those scenes when no dialog was taking place.

Ah, the always-popular translator's notes. AnimEigo is famous for
inserting such notes to accompany the subtitles on some of their
releases, such as "Kimagure Orange Road". I've also seen them on
some fansubs - IIRC, "Kodomo no Omocha" has them, for example.

--
Michael Wojcik michael.wojcik@microfocus.com

Be sure to push the button of the bottom, and push the button of the
settlement page indicated next only once, there is fear of the bottom
rhinoceros multiplex lesson money. -- Sukebe Net

Sponsored Links