This a note from Jack Elhay to Mike Markov.
You will find in MISCLX71 a function INT$. Please believe that there is no
plagiarism - I wrote it before I saw yours. I would, however like your
opinion on it. I was going to write ENG$, SCI$ and FIX$ but your effort
there made mine unnecessary (I hope you don't mind my using them).
I include with this note a BASIC program COPIER which I have transformed
to text in case you wish to assemble my LEX files with a different ID or
tokens. It shows another application of protection against the LIST$ bug.
COPIER will extract a block of BASIC code from one file and transplant
that block to another file. If the destination file does not exist, then
it is created. You don't have to sacrifice the alternate character set.
COPIER uses a couple of my keywords.
In one of my BASIC files I had the line:
820 IF N>75 THEN PRINT CHR$(27)&"&l0L"; ELSE PRINT CHR$(27)&"&l1L";
Now with, ESCSTRLX and TESTLEX2 I can write, instead:
820 PRINT PERF$(TEST(N,75,3));
Don't you think the latter is better?
In case you wonder what value STEP might have, my reason for writing it
is that I can use it in running a program I wrote to do least-squares
curve fitting. STEP allows me to fit least-squares spline models.
E.g.: y=a+b*x+c*STEP(x,s,4)^2 which is the straight line whenever x is
less than or equal to s, i.e. y=a+b*x, but is a parabola forx>s,i.e.
                              y=a+b*x+c*x^2 .
While this can be done by setting an appropriate design matrix, the latter
finishes up having more columns, and worse than that, the problem of 
multicollinearity raises its ugly head.
Some of us in Melbourne feel somewhat isolated from the mainstream of
activity - I guess we depended for too long on what others have done,
and this came to us rather piecemeal from the US and France mainly.
This is changing. There is a growing interest in assembler (I started
in December '86 after complaining for months that it was too hard!)
I started by adapting other people's work and went on to do some of
my own. I was surprised to be told that I was the first from Melbourne
to write an original LEX file. And of those few working on assembler
I was least experienced! I have learnt the hard way( you don't learn
if you don't make mistakes - i.e. MEMORY LOST - HATED MESSAGE!) and
still, every time I write some thing new and assemble it, my heart
doubles its rate until,either the keyword works, or I get some rude
message ( DATA TYPE or more subtle , the wrong answer).
Two people have helped me: Graham Fraser, whom you know, and David
Cameron, who has had some experience with other assemblers and thinks
more easily in HEX than he does in mnemonics. He is working on his
own disassembler.
                 Thanks for taking the time to read this. If you
would like to make any constructive criticism about anything I
have written, I shall be only too pleased to hear from you.
Although I shall be 50 years old next year, (a late starter)
I am finding assembler extremely exciting and often frustrating.
I am consoled, however, that no creative act is ever free of that.
The success at the end and the sense of accomplishment is reward
enough.
       Cheers and Happy Programming,
                                    Jack Elhay,
either c/- PPPM, P.O. Box 512,Ringwood,Victoria,3134,Australia
or 9/494-498 Mitcham Rd,Mitcham,Victoria,3132,Australia
Alternatively- a note in the next swapdisc!