"low-values" 
Author Message
 "low-values"

Quote:



> > That won't work, will it?  The default USAGE in this case will be DISPLAY.
> > Therefore, assigning '0000' will leave in storage the hex values of 'F0F0F0F0'.
> He didn't assign a value of '0000'; but X'0000', which is LOW-VALUES in
> the IBM mainframe world.

Except, Bill, when an alternative alphabet has been defined in the
special-names. In that case x'00' could even be the highest-value!

:(
--

Sorry - but that's wrong (alternative alphabet).  Low-values is low-values regardless of alphabet.  It is infinately low.  A computer doesn't understand infinety so to describe it X'00' is used (x'FF' is infinately high - or high-values).  It doesn't matter which alphabet you use - the machines must use X'00' to represent low-values and X'FF' as high-values.  It is part of their architecture.  Numbers are universal and while high and low values aren't numbers per se and cannot be use in arithmetic they are sorta in the number set of the colating sequence :)  Human minds see this stuff - computers - well - do we really want to know what's on 'their' minds?  (G)

 Sent via Deja.com http://www.*-*-*.com/
 Before you buy.



Tue, 26 Mar 2002 03:00:00 GMT  
 "low-values"

Quote:




> > > That won't work, will it?  The default USAGE in this case will be DISPLAY.
> > > Therefore, assigning '0000' will leave in storage the hex values of 'F0F0F0F0'.
> > He didn't assign a value of '0000'; but X'0000', which is LOW-VALUES in
> > the IBM mainframe world.
> Except, Bill, when an alternative alphabet has been defined in the
> special-names. In that case x'00' could even be the highest-value!

Could be, but that's not the way the smart money bets.

What are the options on alternate alphabets: Cyrillic, Greek, Hebrew?
Are you suggesting someone's going to define another alphabet and make
this universal throughout their files/databases/code? Maybe the NCI or
CIA.

Bill {*filter*} :-)



Tue, 26 Mar 2002 03:00:00 GMT  
 "low-values"
I have heard about (but never actually seen) that some programs have a
"display" zero (X'0F' in IBM mainframe world) setup as low-values.  They then
set all the display numerics followed by the display alphabetics, and then
have all the "other stuff" next the programming collating sequence.

--
Bill Klein
    wmklein <at> ix dot netcom dot com

Quote:




> > > > That won't work, will it?  The default USAGE in this case will be
DISPLAY.
> > > > Therefore, assigning '0000' will leave in storage the hex values of
'F0F0F0F0'.
> > > He didn't assign a value of '0000'; but X'0000', which is LOW-VALUES in
> > > the IBM mainframe world.
> > Except, Bill, when an alternative alphabet has been defined in the
> > special-names. In that case x'00' could even be the highest-value!

> Could be, but that's not the way the smart money bets.

> What are the options on alternate alphabets: Cyrillic, Greek, Hebrew?
> Are you suggesting someone's going to define another alphabet and make
> this universal throughout their files/databases/code? Maybe the NCI or
> CIA.

> Bill {*filter*} :-)



Wed, 27 Mar 2002 03:00:00 GMT  
 "low-values"


Quote:




>> > > That won't work, will it?  The default USAGE in this case will be DISPLAY.
>> > > Therefore, assigning '0000' will leave in storage the hex values of 'F0F0F0F0'.
>> > He didn't assign a value of '0000'; but X'0000', which is LOW-VALUES in
>> > the IBM mainframe world.
>> Except, Bill, when an alternative alphabet has been defined in the
>> special-names. In that case x'00' could even be the highest-value!

>Could be, but that's not the way the smart money bets.

I have to tell that because the guys that invented the ASCII and other
characters sets had a language that did not use special alphabet
charaters like ? or or or and a lot others did not have to worry

about getting them on the proper sort order.
In Portugal and in other countries if we want to have our "Customers"
or our "Articles" sorted by name then we have two options,
1 - Do not allow portuguese characters in the names, and do not allow
    lower case letters.

2 - Define a alphabet in every program and define every indexed file
    to use this collating sequence. (I did use this a lot, although
    it caused some problems when converting file between UNIX/DOS,
    which made me have two copybooks one with the UNIX collating, and
    the other with the Codepage 860, and two compiled versions of the
    programs).

Although I did the second I would not change the order of the
unneeded characters, so it would not change the "low-value" or
"high-value" character.

FF



Wed, 27 Mar 2002 03:00:00 GMT  
 "low-values"

Quote:

> I have heard about (but never actually seen) that some programs have a
> "display" zero (X'0F' in IBM mainframe world) setup as low-values.  They then
> set all the display numerics followed by the display alphabetics, and then
> have all the "other stuff" next the programming collating sequence.

I'm sure X'0F' is a typo & you meant X'F0' (display zero). This
reinforces my point, Bill's heard of such a thing, but hasn't actually
seen it. It boils down to an academic, or even religious (as in "how
many angels can dance on the head of a pin") discussion.

Quote:
> > What are the options on alternate alphabets: Cyrillic, Greek, Hebrew?
> > Are you suggesting someone's going to define another alphabet and make
> > this universal throughout their files/databases/code? Maybe the NCI or
> > CIA.

Mea culpa, "NCI" was supposed to have been the "NSA".

Bill {*filter*}



Wed, 27 Mar 2002 03:00:00 GMT  
 "low-values"

Quote:





etc......

Frederico,

I see you used that dreaded word 'Portuguese' again in your text <G>.

Off-track from what you are talking about, but associated. You are able
to enter 'inflections' over 'a's and 'e's in your text, which we see.
Is this an automatic feature because Windows is picking up your locale
as Iberia, and when you talk about not using them - do you have to
switch this facility off ?

Volker - I assume the same applies with your umlauts and the French with
their 'accents', or our Scandinavian friends, from your recent visit ?

What prompted this - I recall a message a short while back where
somebody was working in Cyprus I believe - and ran into similar alphabet
problems.

Nudging away in the back of my mind - a system is written in N. America,
apart from files for message/dialog texts - what other problems are
their for an 'Anglo' designed system for switching to other languages?
No, I don't want to translate my system to Malay - but we did have a
query from there some three years ago - long before we were prepared to
even consider the enquiry.

If we were talking 'Mother Russia' - I'm sure their national pride would
demand a product be in Cyrillic (?) characters.

Jimmy, Calgary AB



Thu, 28 Mar 2002 03:00:00 GMT  
 "low-values"
Quote:

>Off-track from what you are talking about, but associated. You are able
>to enter 'inflections' over 'a's and 'e's in your text, which we see.
>Is this an automatic feature because Windows is picking up your locale
>as Iberia, and when you talk about not using them - do you have to
>switch this facility off ?

Well, if i want to create the letter "?" I just have to hit "~"
followed by "a". If I dont want to have portuguese characters I just
dont use the 2 character combination.

Quote:

>Volker - I assume the same applies with your umlauts and the French with
>their 'accents', or our Scandinavian friends, from your recent visit ?

It does.

Quote:
>Nudging away in the back of my mind - a system is written in N. America,
>apart from files for message/dialog texts - what other problems are
>their for an 'Anglo' designed system for switching to other languages?

First problem is that the translated message may be to long for the
resulting field, and if changing screens that can be hard to adapt,
normally by changing the source progtrams (or if you just use a screen
manager(like SP2 or RM/PANELS) by changing the screen fields).
If you are translating to languages that use a double byte character
it may be even more dificult, although I had never done this one.

Second you have the sort sequence as I mentioned in a previous post,
although this is normally ignored (to much work to do if you really
need it).

Third and more important of all, all the legal stuff, and way of
doing thing in terms of accounting, personnel and so on. This normally
means customization of certain pieces of the software.

It is easier to convert a general software like a cheques/bank control
system than to convert an fully integrated accounting system.

Hope this helps

FF



Thu, 28 Mar 2002 03:00:00 GMT  
 "low-values"
Quote:

> Frederico,
> Off-track from what you are talking about, but associated. You are able
> to enter 'inflections' over 'a's and 'e's in your text, which we see.
> Is this an automatic feature because Windows is picking up your locale
> as Iberia, and when you talk about not using them - do you have to
> switch this facility off ?

In windows:
- Start
- Settings
- Keyboard
From here on it is sligthly different for NT versus ' 95/' 98:
- Add language english (US/UK whichever you like) and keyboardlayout US.
- Add also english with keyboard leyout US-international.
Now you can switch between these via the start/settings/etc sequence above and
then switching which one is set as default. Faster is to enable the checkbox
"indicator on taskbar" and switch from there (just click on the small indicator
)

Having selected the US/US combi, the characters ' " ~ ` ^ work as normal. Nice
for typing COBOL like MOVE "STRING" TO CHARX.
Having US/International selected, these characters ('"~`^) kind of wait for the
next key. If that is a character which can be combined, it wil. Typing "
followed by e yields ?. Lots of combinations possible of course.
??a????.

And how to get a separate " without switching? Just type the " followed by a
space.

No need to remeber alt-sequences. Works in each and every windowsapp. (Some
people remeber all those obschure Word-alt-control-key-combi's, but these work
in notepad, write, .... all windows-apps.)

I don't know how to get ?'s ?'s etc. with the above method, but I have a great
other little tool for that (as you see :) ), which can freely be downloaded fron
http"//www.whiteants.com, writen by one of those famous dutch programmers!

!

The Frog
--
 ---------------------------------------------------
| H. M. (Huib) Klink   aka      *The COBOL Frog*    |
|-----------------------+---------------------------|

| sMail: P.O. Box 14    |        +31 6 53747091     |
|        3940 AA  Doorn | Visit: Langbroekerdijk A26|
|        The Netherlands|        3947 BH  Langbroek |
 ---------------------------------------------------



Sat, 30 Mar 2002 03:00:00 GMT  
 
 [ 8 post ] 

 Relevant Pages 

1. "low-values"

2. string.join(["Tk 4.2p2", "Python 1.4", "Win32", "free"], "for")

3. "How to measure very low frequency?"...

4. low level "how to scheme"

5. About ''"CooL": low-level macros considered useful

6. BEGIN{want[]={"s1o", "s2o", "s2q", "s3q"}

7. LOW-VALUES as SPACES, HIGH-VALUES as ALL "9"

8. Prop:Order and "null" values

9. Control "value" property

10. "Accept" resets values

11. file drop value disappears on "change"

12. "Lock constant value"

 

 
Powered by phpBB® Forum Software