help! 
Author Message
 help!

My daughter has a homework on APL.She needs a simple sample instruction
in APL and info on where the language is used nowadays (in non-technical
terms). We've been on the net for hours so if there is some patient
person out there who wouldn't mind helping ...please help.

Nenette



Sat, 06 Mar 2004 09:40:48 GMT  
 help!
Recent APL applications in Finland, electricity power distribution area:

1) e-Line
The program collects hourly power load figures (available as EDI messages)
from the power grid, saves them into an MS Access database (equals using the
MS Jet database engine) through the ADO 2.6 COM interface. Then it performs
various calculations on the data and presents the results in the GUI as
tables and charts (screenshot at http://www.enease.fi/taseselv.htm) as well
as outputs into text, Excel and Access report files. In addition it sends
EDI messages to Finngrid (a "central point" in the Finnish power grid) using
XML.

2) QualityReport
kWh-meters with a quality monitoring function collect electricity quality
indicators (like shortages, and more advanced ones). These are put into an
SQL database, from which the figures are selectively read into an APL
"custom made, optimised native file system database". The quality indicators
are finally either presented in the GUI as charts and tables (time series)
or as a 10-page report.

Both are commercial Windows applications, totally about a dozen licenses
distributed so far.

Hope these short examples help... :)
/ Tomas



Quote:
> My daughter has a homework on APL.She needs a simple sample instruction
> in APL and info on where the language is used nowadays (in non-technical
> terms). We've been on the net for hours so if there is some patient
> person out there who wouldn't mind helping ...please help.

> Nenette



Sat, 06 Mar 2004 18:16:47 GMT  
 help!
In addition to what's already been said, don't overlook the fact that APL is
heavily used by Actuaries in the Insurance industry.

APL is also used for scientific work, mathematical computation, financial
modeling, and much more.

Quote:

> My daughter has a homework on APL.She needs a simple sample instruction
> in APL and info on where the language is used nowadays (in non-technical
> terms). We've been on the net for hours so if there is some patient
> person out there who wouldn't mind helping ...please help.

> Nenette



Sun, 07 Mar 2004 02:20:18 GMT  
 help!

Quote:

> In addition to what's already been said, don't overlook the fact that APL is
> heavily used by Actuaries in the Insurance industry.

Given the number of APL users and the number of Actuaries that there are in the
world, could you say what you mean by `heavily used', and what data you base this
judgement on?


Sun, 07 Mar 2004 02:40:28 GMT  
 help!
My experience from STSC through the current time. I do a substantial amount of APL
training and for the most part the audience is actuaries.  At one time there used to
be an acronym joke that the "A" in APL stood for Actuarial (Programming Language).
Quote:


> > In addition to what's already been said, don't overlook the fact that APL is
> > heavily used by Actuaries in the Insurance industry.

> Given the number of APL users and the number of Actuaries that there are in the
> world, could you say what you mean by `heavily used', and what data you base this
> judgement on?



Sun, 07 Mar 2004 03:04:24 GMT  
 help!

Quote:

> My experience from STSC through the current time. I do a substantial amount of APL
> training and for the most part the audience is actuaries.  At one time there used to
> be an acronym joke that the "A" in APL stood for Actuarial (Programming Language).

All well and good, but it still doesn't answer the question.  In my normal use of words
if someone said `Excel is heavily used by financial planners' I would take it to mean that
a _very_ significant proportion of financial planners used Excel (otherwise they surely could
have left out the `heavily'), perhaps even a majority. Yet given the number of Apl users and
the number of Actuaries, I cannot conceive that even 1 actuary in 10 uses APL, and actually
suspect that the number might well be significantly smaller.

I would base this on:
   (1) looking at the sessions at actuarial conferences;
   (2) looking at the curricula of (at least a few) schools well known for
       actuarial studies; and
   (3) looking at the proportion of discussion within the Apl community that is devoted to
       actuarial matters.

My reason for pursuing the question is that if I am looking at the wrong data, or improperly drawing
conclusions from it, I'd like to find out so that I could stop doing it. On the other hand, if I am
not wrong, then saying things like `heavily used' when talking to people who have no other basis for
judgment is misleading at best.



Sun, 07 Mar 2004 03:25:50 GMT  
 help!
As far as I know the sector of industry which at least used APL
was the (life) insurance industry, but nowadays I don't know, I guess
Kevin is a little bit behind ... they're using e.g. C &/or Perl.
Before I retired (3.5 years ago) I've been using APL in the actuarial field
"heavily" (they're still cursing me for that now) - to my knowledge some
companies in the industry REQUIRE(D) (some) knowledge of APL
but in these days ... ?
I think it's turning to J and/or K, while APL has had it's time


Sun, 07 Mar 2004 03:41:36 GMT  
 help!
I would say that Kevin is right, thought this would depend on your
definition of "heavily", as David suggests.

My APL experience started when I was an actuarial assistant at Old
Mutual (largest life assurer in South Africa). I was there until mid
1988, and we used APL2 and VS/APL (VM/CMS on the mainframe) and STSC APL
(DOS on PCs) as well as related products like APLDI. Certainly, in that
company, we used it "heavily".

Other life offices I've worked at:
  Crown Financial Management: didn't use APL, but the less said about
them the better.
  Sun Life of Canada: Heavy use of Sharp APL (MVS), (in Canada) STSC APL
(SunOS)
      and (in UK) Dyalog APL (AIX, SunOS, Solaris, Windoze)
  Commercial Union/CGU/CGNU: Very significant use of STSC (MVS) and
Dyalog (Windoze).

It must be said that SLOC is moving (has moved?) away from APL and doing
a lot of work in C/C++ now. In my view that was a serious mistake.

It is probably worth noting that the actuarial disipline is very wide,
and becoming wider. Many actuaries have little need for advanced
computation, but I believe that a good proportion of those that do
eventually come around to using APL (or getting somebody to do it for
them) once they discover that a spreadsheet has too many limitations. I
know: I was ordered to build a complex model of my employers PHI
liabilities once using Lotus 1-2-3 (on Solaris), which became so large
that we could only run it on the file server and recalcs took over 5
minutes!!

However, this "traditional" actuarial work is just a subset (albeit
pretty specialised) of "financial work", and insurance offices are a
subset of "finacial institutions". Many of these institutions use APL
(not sure if this can be described as "heavily"). Morgan Stanley have
their own variant, and just try to find out who the K clients are.

Getting back to actuaries, if they are not using APL, they {*filter*}y well
should be. Given that much traditional actuarial work involves matrices
and mathematical calculations, APL is the natural language choice.

Quote:


> > My experience from STSC through the current time. I do a substantial amount of APL
> > training and for the most part the audience is actuaries.  At one time there used to
> > be an acronym joke that the "A" in APL stood for Actuarial (Programming Language).

> All well and good, but it still doesn't answer the question.  In my normal use of words
> if someone said `Excel is heavily used by financial planners' I would take it to mean that
> a _very_ significant proportion of financial planners used Excel (otherwise they surely could
> have left out the `heavily'), perhaps even a majority. Yet given the number of Apl users and
> the number of Actuaries, I cannot conceive that even 1 actuary in 10 uses APL, and actually
> suspect that the number might well be significantly smaller.

> I would base this on:
>    (1) looking at the sessions at actuarial conferences;
>    (2) looking at the curricula of (at least a few) schools well known for
>        actuarial studies; and
>    (3) looking at the proportion of discussion within the Apl community that is devoted to
>        actuarial matters.

> My reason for pursuing the question is that if I am looking at the wrong data, or improperly drawing
> conclusions from it, I'd like to find out so that I could stop doing it. On the other hand, if I am
> not wrong, then saying things like `heavily used' when talking to people who have no other basis for
> judgment is misleading at best.

--
-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GO/! d- s++:+ a+ C++(++++) US+++$ P+++ L++ E--- W++ N++ w--- O- V- PS+
PE+ Y+ PGP- t+ 5++ X R* tv+ b+ DI++ D G e(*) h++/-- r+++ y?
------END GEEK CODE BLOCK------

-----------------------------------------------------
Bob Hoekstra:   APL & Unix Consultant
Tele:           +44 (0)1483 771028
                http://www.*-*-*.com/

-----------------------------------------------------



Sun, 07 Mar 2004 03:58:58 GMT  
 help!

Quote:



> > In addition to what's already been said, don't overlook the fact that APL is
> > heavily used by Actuaries in the Insurance industry.

> Given the number of APL users and the number of Actuaries that there are in the
> world, could you say what you mean by `heavily used', and what data you base this
> judgement on?

Although I'm no actuary, let me take a shot at this question.

We ran an item in APL Quote Quad a while back, showing the
results of a survey reported in the CompAct newsletter, from
The Society of Actuaries, which said:

|  Survey Says:
|  APL Gets Top Billing for Usage at Insurance Companies
|
|  CompAct, the newsletter of the computer section of the
|  Society of Actuaries, published a survey on computer usage
|  by its members in May 1997; this is the most recent survey
|  of its type that we have seen.  There were 591 responses
|  to the survey.
|
|  Question 11 was "I use programs written in...".  The top
|  response, at 38%, was "APL".  (Basic was 29%, fortran 27%,
|  COBOL 20%, and others were lower.)
|
|  Question 12 was "I can program in...".  The top four
|  responses were Basic 50%, Fortran 47%, APL 46%, and
|  COBOL 20%.
|
|  Most of the respondents worked for insurance companies.
|
|     --From CompAct, c/o The Society of Actuaries
|        475 Martingale Road, Suite 800, Schaumberg, IL 60173

--Jon McGrew



Sun, 07 Mar 2004 05:05:14 GMT  
 help!

Quote:

> I would say that Kevin is right, thought this would depend on your
> definition of "heavily", as David suggests.

> My APL experience started when I was an actuarial assistant at Old
> Mutual (largest life assurer in South Africa). I was there until mid
> 1988, and we used APL2 and VS/APL (VM/CMS on the mainframe) and STSC APL
> (DOS on PCs) as well as related products like APLDI. Certainly, in that
> company, we used it "heavily".

In companies where it's used, I have no doubt it's used `heavily'. The question,
though, is how much is it used in the industry.

I'm woefully ignorant of South Africa, but have in the back of my head that it's
about 1/6 the size of the US population-wise. Is that about right? Do you
have any idea of how many actuaries there are in SA? I'd guess 1,000 -> 2,000
based on population/GNP comparisons to the US.

Were there any `actuarial APL groups' in SA? If so how big were they? Do you
have any other `community' measures that might be good data for this question.

Quote:
> Other life offices I've worked at:
>   Crown Financial Management: didn't use APL, but the less said about
> them the better.
>   Sun Life of Canada: Heavy use of Sharp APL (MVS), (in Canada) STSC APL
> (SunOS)
>       and (in UK) Dyalog APL (AIX, SunOS, Solaris, Windoze)
>   Commercial Union/CGU/CGNU: Very significant use of STSC (MVS) and
> Dyalog (Windoze).

Given your credentials, I'm not surprised if you have a lot of contact with
companies that happen to do APL. The question, though, is what proportion they
are of all such companies.

Quote:
> It must be said that SLOC is moving (has moved?) away from APL and doing
> a lot of work in C/C++ now. In my view that was a serious mistake.

That may be true, and an interesting question worthy of discussion in a
separate thread.

[snip]

Quote:
> However, this "traditional" actuarial work is just a subset (albeit
> pretty specialised) of "financial work", and insurance offices are a
> subset of "finacial institutions". Many of these institutions use APL
> (not sure if this can be described as "heavily"). Morgan Stanley have
> their own variant, and just try to find out who the K clients are.

True, but if you're expanding the `community' to `financial work' then any
any measure of APL/Total will surely plummet. Think of the _vast_ number of little
financial planning offices scattered all over America with financial planners all
grinding out Excel worksheets. The number of them that have ever even heard
of APL would be vanishingly small.

Quote:
> Getting back to actuaries, if they are not using APL, they {*filter*}y well
> should be. Given that much traditional actuarial work involves matrices
> and mathematical calculations, APL is the natural language choice.

Well, the actuaries who frequent the J-List might disagree. But even then
there is the question of whether _any_ major institution that trains any
quantity of actuaries teaches APL. Do you know of any? If so, how many
people do they train each year?

If I had to estimate the number of actuaries I guess I'd put it in the
50,000 -> 100,000 range. If I had to estimate the number of active APL
users, I'd put it at way less than 10,000. So, given my estimates, even if
_every_ APL user were an actuary, it would still be less than 1/5 of the
actuaries under assumptions that wildly overestimate the figure. My real
`guess' would be < 1/100, but I emphasize that this is only because
so far no one has suggested any reasonable data that might make me think
otherwise.

So, from a data standpoint, I'm still missing the argument that leads you
to conclude that `Kevin is right' and that APL is `heavily used by Actuaries'.



Sun, 07 Mar 2004 05:49:30 GMT  
 help!

Quote:

> So, from a data standpoint, I'm still missing the argument that leads you
> to conclude that `Kevin is right' and that APL is `heavily used by Actuaries'.

So?  

I had a similar problem with you a while back, David, in that you seem
to think you are owed a proof of any assertion and owed the research
required of a person to make it.  This forum is not a degree conferring
committee upon which you sit.  Repetition of something known but not
personally proven is admissible.  Disproof is incumbent on the
challenger.

Bob
--

"Things should be described as simply as possible, but no simpler."

                                             A. Einstein

////////////////////////////////////////\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

 To contribute your unused processor cycles to the fight against cancer:

     http://www.intel.com/cure

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\///////////////////////////////////////



Sun, 07 Mar 2004 08:10:33 GMT  
 help!
_Very_ dangerous this actually bothering to find out some interesting
facts. And your taking the time to point the sources out is very helpful,
as there is some later information which may allow some to draw useful
conclusions about trends from 1997 thru 1999.

Quote:

> Although I'm no actuary, let me take a shot at this question.

> We ran an item in APL Quote Quad a while back, showing the
> results of a survey reported in the CompAct newsletter, from
> The Society of Actuaries, which said:

I would imagine that CompAct is the `newsletter' of the SOA that
is `computational' in charter. Do you have any idea of what percentage of the
membership gets this? I don't know enough about the business to know, but I
would guess that it would probably go to only a subset of the actuaries.
If anyone knows, I'd be interested in what the numbers are. The reason this
is relevant is that I suppose there are actuaries that may _not_ be interested
in computation (like lots of my mathematician friends) and knowing what the
`population' is helps me understand the numbers.

Quote:
> |  Survey Says:
> |  APL Gets Top Billing for Usage at Insurance Companies
> |
> |  CompAct, the newsletter of the computer section of the
> |  Society of Actuaries, published a survey on computer usage
> |  by its members in May 1997; this is the most recent survey
> |  of its type that we have seen.  There were 591 responses
> |  to the survey.

CompAct did another survey in 1999. It is reported on their Web
site. I have not yet found a detailed report, so I don't know
how it answers the questions that correspond to the questions you
report.

But it does suggest rather different conclusions, at least if
one looks towards the future.

Quote:
> |  Question 11 was "I use programs written in...".  The top
> |  response, at 38%, was "APL".  (Basic was 29%, Fortran 27%,
> |  COBOL 20%, and others were lower.)

The percentages for Fortran and Cobol strike me as startling. I would _not_
have been surprised by these numbers if they were from 1977, but I find
them startling for 1997. Does this suggest anything about the `age' of
actuarial systems?

Quote:
> |
> |  Question 12 was "I can program in...".  The top four
> |  responses were Basic 50%, Fortran 47%, APL 46%, and
> |  COBOL 20%.
> |
> |  Most of the respondents worked for insurance companies.
> |
> |     --From CompAct, c/o The Society of Actuaries
> |        475 Martingale Road, Suite 800, Schaumberg, IL 60173

> --Jon McGrew

The current figures that your pointing me to the source allow me
to add are from their 1999 survey where 20 of the 66 schools they
questioned responded to their survey. They indicate that the
`Percentage Distribution of the programming languages and  other
tools actually taken at the responding schools are:

92% Excel, Lotus Other Spreadsheet
39% C, C++                          
30% Mathematica, Maple              
30% Basic, Visual Basic            
20% Java                            
16% SAS, S-Plus                    
13% Pascal                          
12% Fortran                        
 9% APL, J                          
 4% Lisp                              
 4% Cobol                            
 3% ADA                              
 3% TAS, PTS                        
 2% Advanced Stochastic Languages    

If I understand the supporting data, it indicates that the APL/J
percentage comes from 3 of the 20 schools that report data. Although
there is no direct explanation of this data that I have so far seen,
it is my belief that the percentages are based on `inference' from
`survey facts' like `25% or our students take a course involving APL'
rather than from actual head counts.

It looks to me like it might be possible to make some inferences about
the age of the population that responded to the 1997 survey, or about
the distribution of legacy systems in organizations, but I haven't yet
seen enough data yet to want to do any of that.



Sun, 07 Mar 2004 09:23:37 GMT  
 help!
Another group of "heavy" APL users...left handed Canadians who own skills saws and
like to holiday in the mountains.

Let's see, David, can you qualify and quantify how many left handed Canadians
there are (or should be make it left handed versus ambidexterous?) and how many
skill saws there are in Canada and who owns them, etc., etc.  YIKES...do you have
anything else to do with your time?

Quote:

> _Very_ dangerous this actually bothering to find out some interesting
> facts. And your taking the time to point the sources out is very helpful,
> as there is some later information which may allow some to draw useful
> conclusions about trends from 1997 thru 1999.


> > Although I'm no actuary, let me take a shot at this question.

> > We ran an item in APL Quote Quad a while back, showing the
> > results of a survey reported in the CompAct newsletter, from
> > The Society of Actuaries, which said:

> I would imagine that CompAct is the `newsletter' of the SOA that
> is `computational' in charter. Do you have any idea of what percentage of the
> membership gets this? I don't know enough about the business to know, but I
> would guess that it would probably go to only a subset of the actuaries.
> If anyone knows, I'd be interested in what the numbers are. The reason this
> is relevant is that I suppose there are actuaries that may _not_ be interested
> in computation (like lots of my mathematician friends) and knowing what the
> `population' is helps me understand the numbers.

> > |  Survey Says:
> > |  APL Gets Top Billing for Usage at Insurance Companies
> > |
> > |  CompAct, the newsletter of the computer section of the
> > |  Society of Actuaries, published a survey on computer usage
> > |  by its members in May 1997; this is the most recent survey
> > |  of its type that we have seen.  There were 591 responses
> > |  to the survey.

> CompAct did another survey in 1999. It is reported on their Web
> site. I have not yet found a detailed report, so I don't know
> how it answers the questions that correspond to the questions you
> report.

> But it does suggest rather different conclusions, at least if
> one looks towards the future.

> > |  Question 11 was "I use programs written in...".  The top
> > |  response, at 38%, was "APL".  (Basic was 29%, Fortran 27%,
> > |  COBOL 20%, and others were lower.)

> The percentages for Fortran and Cobol strike me as startling. I would _not_
> have been surprised by these numbers if they were from 1977, but I find
> them startling for 1997. Does this suggest anything about the `age' of
> actuarial systems?

> > |
> > |  Question 12 was "I can program in...".  The top four
> > |  responses were Basic 50%, Fortran 47%, APL 46%, and
> > |  COBOL 20%.
> > |
> > |  Most of the respondents worked for insurance companies.
> > |
> > |     --From CompAct, c/o The Society of Actuaries
> > |        475 Martingale Road, Suite 800, Schaumberg, IL 60173

> > --Jon McGrew

> The current figures that your pointing me to the source allow me
> to add are from their 1999 survey where 20 of the 66 schools they
> questioned responded to their survey. They indicate that the
> `Percentage Distribution of the programming languages and  other
> tools actually taken at the responding schools are:

> 92% Excel, Lotus Other Spreadsheet
> 39% C, C++
> 30% Mathematica, Maple
> 30% Basic, Visual Basic
> 20% Java
> 16% SAS, S-Plus
> 13% Pascal
> 12% Fortran
>  9% APL, J
>  4% Lisp
>  4% Cobol
>  3% ADA
>  3% TAS, PTS
>  2% Advanced Stochastic Languages

> If I understand the supporting data, it indicates that the APL/J
> percentage comes from 3 of the 20 schools that report data. Although
> there is no direct explanation of this data that I have so far seen,
> it is my belief that the percentages are based on `inference' from
> `survey facts' like `25% or our students take a course involving APL'
> rather than from actual head counts.

> It looks to me like it might be possible to make some inferences about
> the age of the population that responded to the 1997 survey, or about
> the distribution of legacy systems in organizations, but I haven't yet
> seen enough data yet to want to do any of that.



Sun, 07 Mar 2004 09:39:20 GMT  
 help!

Quote:

> Another group of "heavy" APL users...left handed Canadians who own skills saws and
> like to holiday in the mountains.

Really? Can't imagine it's interesting, or useful to know that. But apparently
you think it's worth mentioning, I'll put it in the file.

Quote:
> Let's see, David, can you qualify and quantify how many left handed Canadians
> there are (or should be make it left handed versus ambidexterous?) and how many
> skill saws there are in Canada and who owns them, etc., etc.  

No. I know nothing about them. But then I don't make assertions about them either.

Quote:
> YIKES...do you have anything else to do with your time?

Sure. Lots. But then I work hard, so I get a lot done.


Sun, 07 Mar 2004 09:57:30 GMT  
 help!

Quote:

> ...

> So, from a data standpoint, I'm still missing the argument that leads you
> to conclude that `Kevin is right' and that APL is `heavily used by Actuaries'.

The term "heavily used by actuaries" is ambiguous and can mean various
things.  David Ness interprets it to mean that a large percentage of
actuaries use APL and wants proof of this.  Another interpretation could
be a relatively large proportion of actuaries use APL compared to other
professions.  Kevin's experience has been that a large number of his APL
students were actuaries.  This is a common problem with "proving" things
with anecdotal information.  True stories make good arguments, but often
fail to take into account the relevance of the sample to the question at
hand.

Steve
--

"When you lie wounded on the Afghan plain, and Indian women come to pick
your
remains, just roll on your rifle and blow out your brains, and go to
your God
like a soldier." - Rudyard Kipling



Sun, 07 Mar 2004 10:08:46 GMT  
 
 [ 28 post ]  Go to page: [1] [2]

 Relevant Pages 

1. help! f90.help help help help

2. ***HELP***HELP***NEED INFORMATION***HELP***HELP

3. HELP HELP HELP HELP

4. HELP HELP HELP HELP

5. Ord Function HELP Please HELP HELP HELP

6. help help help help!!!!!!!!!!!

7. (HELP (HELP (HELP (HELP))))

8. HELP: HELP: HELP: HELP: Online-manual on Expect

9. Help Help Help

10. TopSpeed - ODBC 3.1???? HELP HELP HELP

11. HELP - HELP - HELP

12. HELP HELP HELP - round function error

 

 
Powered by phpBB® Forum Software