Caselessness...

Caselessness...

[FGD]

This is a distant followup to another thread. I don't think Ada's case 
insensitiveness is a problem, though I wonder... Ada 95 was designed to 
support characters from many different natural languages. Couldn't there 
be a "loophole" here, couldn't a disgruntled C programmer define a 
CaseSensitiveEnglish language wherein upper and lower case analogues are 
distinguished just like 'e' and '�' are distinguished in French?

Frank Dorais

Re: Caselessness...

[Marin David Condic]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1700 bytes --]

If someone is really convinced that it is a problem, then we'd have another
instance where the language can be "improved" with a preprocessor. Build
something that collects up the identifiers and creates unique IDs for them
and mangles up some case-insensitive names for legal Ada output. I just have
my doubts that this is such a horrendous problem for Ada (I'd consider it an
advantage!) that if only it had been made case sensitive, C/C++/Java
programmers would be flocking to it by the millions.

It might even be marketable for those who want to work with a case-sensitive
version of Ada. It would need only be smart enough to distinguish between
regular Ada files and Case Sensitive Ada(tm). Clearly it is something a lot
of people would prefer. Would they prefer it enough to live with a
non-standard Ada variant & use a preprocessor? (Better they decide what they
are willing to live with than break all the existing Ada code that is out
there! :-)

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"FGD" <presbeis@look.ca> wrote in message news:3BCB2FDA.8060807@look.ca...
> This is a distant followup to another thread. I don't think Ada's case
> insensitiveness is a problem, though I wonder... Ada 95 was designed to
> support characters from many different natural languages. Couldn't there
> be a "loophole" here, couldn't a disgruntled C programmer define a
> CaseSensitiveEnglish language wherein upper and lower case analogues are
> distinguished just like 'e' and '�' are distinguished in French?
>
> Frank Dorais
>

Re: Caselessness...

[FGD]

I think you're still reacting to that other thread which I wanted to 
avoid by starting a new thread...

My question is technical: Is the "loophole" I mentionned possible within 
the Ada language specs? (I don't really care if it can be done with a 
preprocessor, or even if it is desirable to do so with or without a 
preprocessor...)

Marin David Condic wrote:

> If someone is really convinced that it is a problem, then we'd have another
> instance where the language can be "improved" with a preprocessor.


Many Ada compilers provide some preprocessor support this does not go 
against Ada design, though Ada doesn't require it and mos often doesn't 
need it.

> Build something that collects up the identifiers and creates unique IDs for them
> and mangles up some case-insensitive names for legal Ada output. I just have
> my doubts that this is such a horrendous problem for Ada (I'd consider it an
> advantage!) that if only it had been made case sensitive, C/C++/Java
> programmers would be flocking to it by the millions.


There's already plenty of that and for legitimate reasons, e.g. see the 
CORBA to Ada language mapping. But it is irrelevant to my question.


> It might even be marketable for those who want to work with a case-sensitive
> version of Ada. It would need only be smart enough to distinguish between
> regular Ada files and Case Sensitive Ada(tm). Clearly it is something a lot
> of people would prefer.


It might be marketable, but is it acceptable within the current specs?

> Would they prefer it enough to live with a
> non-standard Ada variant & use a preprocessor? 


Wherein the specs do you see that it is against standard, or even 
undesirable, to work with a preprocessor?

> (Better they decide what they are willing to live with than break all 

> the existing Ada code that is out there! :-)

AFAIK all Ada compilers are required to support at least ASCII so all 
English based Ada code out there should compile with any validated 
compiler. But the ada specs explicietly allow for other language 
support---so if we allow support for Greek and Russian why don't we 
allow support for CaseSensitiveEnglishForDisgruntledCProgrammers(tm)? ;-)

Re: Caselessness...

[Marin David Condic]

My best guess would be that it would amount to a representation issue. In
the same way that you could, for example, represent the reserved words in an
editor with a different language, you could probably represent identifiers
in some alternate way. The trick would be that you probably have to go both
directions - you have to be able to read the validation suite and compile
it, which means your representation issue comes between the viewer and the
underlying ASCII character set. So imagine an editor that read the ASCII
text and changed the representation for the viewer in some way. Likewise,
what the viewer types is translated to some underlying ASCII and compiled.
So in a sense, my preprocessor suggestion is kind of an answer - except
you'd need something that transformed in both directions so you could
properly handle the validation suite. Remember that the validation suite has
probably got tests in it to insure that character case doesn't matter, since
this is in the standard.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"FGD" <presbeis@look.ca> wrote in message news:3BCB3E2A.1070001@look.ca...
> I think you're still reacting to that other thread which I wanted to
> avoid by starting a new thread...
>
> My question is technical: Is the "loophole" I mentionned possible within
> the Ada language specs? (I don't really care if it can be done with a
> preprocessor, or even if it is desirable to do so with or without a
> preprocessor...)
>

Re: Caselessness...

[David Starner]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 805 bytes --]

On Mon, 15 Oct 2001 14:50:02 -0400, FGD <presbeis@look.ca> wrote:
> This is a distant followup to another thread. I don't think Ada's case 
> insensitiveness is a problem, though I wonder... Ada 95 was designed to 
> support characters from many different natural languages. Couldn't there 
> be a "loophole" here, couldn't a disgruntled C programmer define a 
> CaseSensitiveEnglish language wherein upper and lower case analogues are 
> distinguished just like 'e' and '�' are distinguished in French?

Huh? How would he define this hypothetical thing?

-- 
David Starner - dstarner98@aasaa.ofe.org
Pointless website: http://dvdeug.dhis.org
"I saw a daemon stare into my face, and an angel touch my breast; each 
one softly calls my name . . . the daemon scares me less."
- "Disciple", Stuart Davis

Re: Caselessness...

[Ted Dennison]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1423 bytes --]

In article <3BCB2FDA.8060807@look.ca>, FGD says...
>
>This is a distant followup to another thread. I don't think Ada's case 
>insensitiveness is a problem, though I wonder... Ada 95 was designed to 
>support characters from many different natural languages. Couldn't there 
>be a "loophole" here, couldn't a disgruntled C programmer define a 
>CaseSensitiveEnglish language wherein upper and lower case analogues are 
>distinguished just like 'e' and '�' are distinguished in French?

Assuming it could still compile old code, then that won't change the fact that
"begin" and "BEGIN" are both the same keyword.

This decision was made *for a reason*. Its perfectly legitimate for a beginner
to wonder why certain desicions were made with the language. However, I don't
think its at all legitimate for Ada novices to try to come in here and
repeatedly argue against every single decision the language designers made,
which is the situation this NG seems to be in now. I really gets to be
exhausting trying to explain the rationale for things over and over again to
people who come in here thinking the've single-handedly solved the languages
"problem", and aren't interested in hearing why things were put the way they
are.

---
T.E.D.    homepage   - http://www.telepath.com/dennison/Ted/TED.html

No trees were killed in the sending of this message. 
However a large number of electrons were terribly inconvenienced.

Re: Caselessness...

[Jeffrey Carter]

Ted Dennison wrote:
> 
> This decision was made *for a reason*. Its perfectly legitimate for a beginner
> to wonder why certain desicions were made with the language. However, I don't
> think its at all legitimate for Ada novices to try to come in here and
> repeatedly argue against every single decision the language designers made,
> which is the situation this NG seems to be in now. I really gets to be
> exhausting trying to explain the rationale for things over and over again to
> people who come in here thinking the've single-handedly solved the languages
> "problem", and aren't interested in hearing why things were put the way they
> are.

It would be useful to be able to reference this reason. I have not found
any justification for case insensitivity in the Ada Rationale or the
Ada-83 Rationale. Maybe I missed it. Another possibility is that this
was a requirement imposed on the language design, in which case the
justification would probably be found in Steelman.

-- 
Jeffrey Carter

Re: Caselessness...

[Al Christians]

Wasn't there originally a requirement to target some machines that
didn't support upper and lower case characters?  For example, the 
typical printer chains that were used on mainframe printers in the early 
1980's didn't have lower case characters on them at all. Case sensitive
would be a major debacle if a machine like that was used to print
code listings.


Al



Jeffrey Carter wrote:
> 
> Ted Dennison wrote:
> >
> > This decision was made *for a reason*. Its perfectly legitimate for a beginner
> > to wonder why certain desicions were made with the language. However, I don't
> > think its at all legitimate for Ada novices to try to come in here and
> > repeatedly argue against every single decision the language designers made,
> > which is the situation this NG seems to be in now. I really gets to be
> > exhausting trying to explain the rationale for things over and over again to
> > people who come in here thinking the've single-handedly solved the languages
> > "problem", and aren't interested in hearing why things were put the way they
> > are.
> 
> It would be useful to be able to reference this reason. I have not found
> any justification for case insensitivity in the Ada Rationale or the
> Ada-83 Rationale. Maybe I missed it. Another possibility is that this
> was a requirement imposed on the language design, in which case the
> justification would probably be found in Steelman.
> 
> --
> Jeffrey Carter

Re: Caselessness...

[Peter Hend�n]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 904 bytes --]

Al Christians wrote:
> Wasn't there originally a requirement to target some machines that
> didn't support upper and lower case characters?  For example, the
> typical printer chains that were used on mainframe printers in the early
> 1980's didn't have lower case characters on them at all. Case sensitive
> would be a major debacle if a machine like that was used to print
> code listings.
As I recall from a seminar with Ichbiah, Barnes and someone else (it
should have been Firth, but he was replaced for some reason), this
was one of the reasons. I also seem to remember that discriminating
names depending on case was out of the question because it was
error prone.

I know I have notes of this somewhere, but going back 20 years is
iffy at best.

Regards,
Peter H.

--
Peter Hend�n           http://www.algonet.se/~phenden
ICQ: 14672398
Teknisk Dokumentation AB          http://www.tdab.com

Re: Caselessness...

[Jeffrey Carter]

Al Christians wrote:
> 
> Wasn't there originally a requirement to target some machines that
> didn't support upper and lower case characters?  For example, the
> typical printer chains that were used on mainframe printers in the early
> 1980's didn't have lower case characters on them at all. Case sensitive
> would be a major debacle if a machine like that was used to print
> code listings.

There was a requirement to allow source to be represented on machines
with a limited character set; for example, '%' could be used instead of
'"', and ':' instead of '#' in based numeric literals; see Annex J and
J.2 in particular. However, this did not make programs using '"'
illegal, while a program with 2 identifiers identical except for case is
illegal, so this doesn't appear to be the reason for case insensitivity.

-- 
Jeffrey Carter

Re: Caselessness...

[Al Christians]

Well, if you were trying to port code from a case-sensitive machine
to one that could not represent both cases, the code could break if
the language was case-sensitive.  Back when Ada 83 was being developed,
there were many programs still being made machine-readable by keypunch 
machines.  Those machines were not built to represent more than one 
case, IIRC.  It would have been a bad problem to try to put any code
that took advantage of case sensitiveness through those machines, so 
case sensitiveness  would have been a big hit to portability.

Al

Jeffrey Carter wrote:
> 
> Al Christians wrote:
> >
> > Wasn't there originally a requirement to target some machines that
> > didn't support upper and lower case characters?  For example, the
> > typical printer chains that were used on mainframe printers in the early
> > 1980's didn't have lower case characters on them at all. Case sensitive
> > would be a major debacle if a machine like that was used to print
> > code listings.
> 
> There was a requirement to allow source to be represented on machines
> with a limited character set; for example, '%' could be used instead of
> '"', and ':' instead of '#' in based numeric literals; see Annex J and
> J.2 in particular. However, this did not make programs using '"'
> illegal, while a program with 2 identifiers identical except for case is
> illegal, so this doesn't appear to be the reason for case insensitivity.
> 
> --
> Jeffrey Carter

Re: Caselessness...

[David Brown]

FGD <presbeis@look.ca> wrote:

> This is a distant followup to another thread. I don't think Ada's case 
> insensitiveness is a problem, though I wonder... Ada 95 was designed to 
> support characters from many different natural languages. Couldn't there 
> be a "loophole" here, couldn't a disgruntled C programmer define a 
> CaseSensitiveEnglish language wherein upper and lower case analogues are 
> distinguished just like 'e' and '?' are distinguished in French?

AFIK, the English language only has one pair of words that can be
distinguished by case: August (the month), and august (marked by
majestic dignity or grandeur).  Since one is a noun, and the other is an
adjective, I would have a hard time seeing them used in a conflicting
way.  The adjective variant is not used frequently in English anyway.

Besides that, the adjective at the start of a sentence is still the
adjective.  English is ambiguous in many other ways.

Identifiers generally aren't capitalized according to English rules
anyway, but according to style guidelines.

Dave Brown

Re: Caselessness...

[David Starner]

On Wed, 17 Oct 2001 04:56:55 GMT, David Brown <davidb-cla@davidb.org> wrote:
> AFIK, the English language only has one pair of words that can be
> distinguished by case: August (the month), and august (marked by
> majestic dignity or grandeur).  

Polish/polish (both adjectives). (I've heard Polish/polish quoted; I've
never seen August/august before.) Hope (Faith, Charity, Rose, etc.) when
used as names versus their ordinary uses.

Not arguing agianst your points; just making a minor correction.

-- 
David Starner - dstarner98@aasaa.ofe.org
Pointless website: http://dvdeug.dhis.org
"I saw a daemon stare into my face, and an angel touch my breast; each 
one softly calls my name . . . the daemon scares me less."
- "Disciple", Stuart Davis

Re: Caselessness...

[Ted Dennison]

In article <rc8z7.5446$SU2.573487@newsread1.prod.itd.earthlink.net>, David Brown
says...
>
>AFIK, the English language only has one pair of words that can be
>distinguished by case: August (the month), and august (marked by
>majestic dignity or grandeur).  Since one is a noun, and the other is an
>adjective, I would have a hard time seeing them used in a conflicting
>way.  The adjective variant is not used frequently in English anyway.

If you throw in acronyms, there are oodles of them. For instance, mad (angry or
insane) and MAD (the defensive philosophy of mutually-assured destruction,
perhaps also insane but often effective). Of course the relevant example here is
ADA (acronym for American Dental Association, Americans with Disabilities Act,
and many many more) and Ada (the name of many people, a city in Oklahoma, and
the world's best programming language).

---
T.E.D.    homepage   - http://www.telepath.com/dennison/Ted/TED.html

No trees were killed in the sending of this message. 
However a large number of electrons were terribly inconvenienced.

Re: Caselessness...

[Philip Anderson]

Ted Dennison wrote:
> 
> In article <rc8z7.5446$SU2.573487@newsread1.prod.itd.earthlink.net>, David Brown
> says...
> >
> >AFIK, the English language only has one pair of words that can be
> >distinguished by case: August (the month), and august (marked by
> >majestic dignity or grandeur).  Since one is a noun, and the other is an
> >adjective, I would have a hard time seeing them used in a conflicting
> >way.  The adjective variant is not used frequently in English anyway.

August company assembles?

March triumph?


But there are plenty of syntactic and semantic ambiguities which can't
be solved by capitalisation, eg "flying planes can be dangerous".


> If you throw in acronyms, there are oodles of them. For instance, mad (angry or
> insane) and MAD (the defensive philosophy of mutually-assured destruction,
> perhaps also insane but often effective). Of course the relevant example here is
> ADA (acronym for American Dental Association, Americans with Disabilities Act,
> and many many more) and Ada (the name of many people, a city in Oklahoma, and
> the world's best programming language).

So writing ADA (or Ada) is still ambiguous.  That's the way English (and
every other language) is, there is always some ambiguity to be solved by
context (and redundancy).


-- 
hwyl/cheers,
Philip Anderson
Alenia Marconi Systems
Cwmbr�n, Cymru/Wales

Re: Caselessness...

[Ted Dennison]

In article <3BCDA863.1C27DDC4@amsjv.com>, Philip Anderson says...
>
>Ted Dennison wrote:
>> 
>> perhaps also insane but often effective). Of course the relevant example 
>> here is ADA (acronym for American Dental Association, Americans with 
>> Disabilities Act, and many many more) and Ada (the name of many people, a 
>> city in Oklahoma, and the world's best programming language).
>
>So writing ADA (or Ada) is still ambiguous.  That's the way English (and
>every other language) is, there is always some ambiguity to be solved by
>context (and redundancy).

True. However, ambiguity isn't the issue here; its correctness. The word "mad"
many have several meanings, but none of them are "mutually-assured destruction".
The word "ADA" also has several meanings, but none of them are a programming
language.

---
T.E.D.    homepage   - http://www.telepath.com/dennison/Ted/TED.html

No trees were killed in the sending of this message. 
However a large number of electrons were terribly inconvenienced.

Re: Caselessness...

[Darren New]

David Brown wrote:
> AFIK, the English language only has one pair of words that can be
> distinguished by case: August (the month), and august (marked by
> majestic dignity or grandeur). 

Unlike "Polish" and "polish", August and august both come from the same
root word, as well.

-- 
Darren New 
San Diego, CA, USA (PST). Cryptokeys on demand.
                   Who is this Dr. Ibid anyway, 
                  and how does he know so much?