World: r3wp

I can't help being disappointed (although I trust the team to do 
great work)

I need something being tested on older versions of REBOL. So if some 
of you could run this little piece of code and report, what the end 
value is, and what version of REBOL under what OS and on what hardware. 
The piece of code:

x: 1.0 while [(1.0 + (x / 2.0)) > 1.0] [x: x / 2.0]

On my G4 iBook, system/version is 2.7.5.2.4 and I get: 3.5527136788005E-15

REBOL/Core 2.5.0.3.1

>> x: 1.0 while [(1.0 + (x / 2.0)) > 1.0] [x: x / 2.0]
== 3.5527136788005E-15

XP P 4 2.66 1 GB

REBOL/View 1.3.1.3.1 17-Jun-2005 Core 2.6.0 under Win2000 (not sure 
which CPU):
1.77635683940025E-15

REBOL/Core 2.5.6.7.1 under FreeBSD i386:
3.5527136788005E-15

>> system/version ;View 1.3.2.3.1 5-Dec-2005 Core 2.6.3
== 1.3.2.3.1
>> x: 1.0 while [(1.0 + (x / 2.0)) > 1.0] [x: x / 2.0]
== 1.77635683940025E-15
;Athlon64 2.0GHz/w32 5.1

REBOL/View 1.3.2.3.1 5-Dec-2005 Core 2.6.3
Copyright 2000-2005 REBOL Technologies.  All rights reserved.
REBOL is a trademark of REBOL Technologies. WWW.REBOL.COM
>> x: 1.0 while [(1.0 + (x / 2.0)) > 1.0] [x: x / 2.0]
== 1.77635683940025E-15

Microsoft Windows XP Home Edition Version 2002 Service Pack 2   Compaq 
Presario AMD Athlon XP 2200+  1.80 GHz  1.00 GB of RAMS

James, not sure yet. Probably not, as I need unicode support.

REBOL/Core 2.5.6.3.1 / win XP / AMD-TL50 
3.5527136788005E-15

REBOL/View 1.3.1.3.1 17-Jun-2005 Core 2.6.0  -- same op sys and machine 
as above:
1.77635683940025E-15

Windows XP Pro,  Prozessor	x86 Family 15 Model 2 Stepping 9 GenuineIntel 
~2807 Mhz

REBOL/Command 2.6.2.3.1
1.77635683940025E-15
REBOL/Pro 2.6.2.3.1
1.77635683940025E-15
REBOL/Core 2.6.0.3.1
1.77635683940025E-15
REBOL/Pro 2.5.125.3.1
1.77635683940025E-15

REBOL/Core 2.5.0.3.1
3.5527136788005E-15

REBOL/Pro 2.5.8.3.1
8.88178419700125E-16

REBOL/Core 2.3.0.3.1
false

REBOL/Core 2.2.7.3.1
false

Geomol: you discovered "rounded equality", unfortunately, there were 
some adjustments to it recently (some made by mistake, some intentional)

have you got any reasons why to prefer same? 1 1.0 == true or same? 
1 1.0 == false?

(question is for any rebol user and mainly for Rebol, i.e. the future)

...for Rebol3... is what I wanted to mention above

Okay, I don't think, I need more tests on my little routine. The 
8.88.....E-16 result from sqlab was interesting! :-) Else all other 
results were 1.77....E-15 and 3.55.....E-15.


Ladislav, I think, I understand the situation and the problem now. 
That's why I made this investigation. There is another way to handle 
this, where it's up to the programmer, which of the two situations
same? 1 1.0 == true or same? 1 1.0 == false?

should hold. See my large post in the "Bugs" group in "R3-Alpha" 
world.

Thanks to everyone doing the test!

A little update from Alpha testing. Since last time, this happened:

- POWER can now handle negative number and exponent

- Some bugs fixed regarding: money!, path, VID crash, change/part, 
read, function and closure recursion crash, compose/deep
- New dictionary! datatype (replacing hash!)
- A lot is going on in the graphics, VID and DRAW groups
- Ongoing work to get the test methods to perfection

We're now on Alpha 49.

Carl asks about better name for dictionary!, which is a bit long. 
I think there is only one alternative, if we want the name to be 
on pair with other languages - dict! ... the thing is, that we don't 
use abbreviations in REBOL most of the time, but we have func too, 
so why not dict! ?

and I don't like percents. I don't want to open that discussion here, 
because I already seen some discussion on it, and while it might 
seem trivial, it is not :-) But generally I refuse result which is 
different from what I got from calculator. So basically how following 
could be valid escapes my basic school knowledge:

12.3 * 110% = 13.4


Of course I would expect 12.3 + (10% from the base (12.3)) = 13.53, 
which is returned also by my Windows calculator. Even if I think 
about 110% as of 1.1, still 12.3 * 1.1 = 13.53.

IMO there is a bug in the doc :-)

that's a docs problem, let me fix it...
r3 works as your calculator, Pekr :)
>> 12.3 * 110%
== 13.53

Also remember that main purpose of percent is to enhance sematics 
in dialects and not to work as a calculator. But you're free to write 
your own calculator dialect.

I just read about BCD representation and money datatype. What escapes 
my understanding is, why BCD is internally tied to just money area? 
I really don't like it.

Coulldn't I just choose some mode for decimal datatype to work as 
BCD?

E.g. - just recently in one of our systems we tried to solve weigh 
unit conversions, where we needed precision on milligrams. If my 
understanding is correct, it is where I should use money. Now how 
is $ unit usefull for me here?

Maybe best to think of the $ as a BCD specifer rather than specifically 
a money one. 
   to-bcd: : to-money
  for all other uses :-)

yes, I would come up with some new name. My question is - did anyone 
use money? I never used it in my scripts for e.g.

just recently dictionary! datatype was renamed to map! datatype ...

Using $ as a BCD specifier wouldn't be so bad if 'print ignored it. 

>>print join weight ["mg"]
== $123.45678mg

some time ago we thought about allowing other chars for money. Or 
something like USD$123.123 ... the question is, if there is some 
better char to be used for amount of various units?

$123.45678mg .... That is a drawback, Peter. Will format help?
And it may still be negotiable in R3:
http://www.rebol.net/r3blogs/0092.html

format is not enough. I don't know why we have datatype limited to 
money? Why not kg, tonnes, other units?

the trouble is, how to make such datatype look like in rebol, and 
 not complicate low level parser

we have already #[none] form etc., so maybe [kg]1234.5678, or 1234.5678[kg] 
:-)

With 64-bit decimals, you have 15 digits precision, and it's faster. 
You can specify up to a million ton, and still have milligrams. Is 
that enough for your task?

Example: 123'456'789.123'465 is a valid decimal! with full precision.

Geomol - yes, it is. And is it guaranteed that the last digits will 
be always precise?

That's one of the things, I'm investigating for the alpha-release. 
If all your numbers are within that limit, it should be ok. But if 
you e.g. add 2 numbers like: 123'456'789'012'345.0 + 123'456'789.012'345, 
you loose presicion in the smaller number.

The precision for 64-bit IEEE decimals is 15.95 decimal digits. You 
can have results up to 17 digits (it's a parameter to be able to 
change). It's default set to 15 to always give correct result. In 
5% of the cases, the 16th digit will be wrong. See e.g.: http://babbage.cs.qc.edu/courses/cs341/IEEE-754references.html

Work is done to get around problems like:
>> (0.2 - 0.1) = 0.1
== true
>> (1.2 - 1.1) = 0.1
== false
(This is also a the results in R2.)

And in other languages like C btw.

well, look - we've got BCD for the money, right? so there is the 
solution for those who need it. But i would vote for another kind 
of expressing units, more general. IMO money! is the most useless 
datatype in REBOL. My proposition really is 123'456.123[unit], where 
'unit would be of no meaning to not complicate things - simply 123[USD] 
+ 123[EUR] would be 246[USD] - the first occurance of the unit.

I can even imagine unit conversion table, so that we could get even 
USD + EUR result .... and if expression would contain some non transferrable 
units, e.g. USD + kg, then error would be raised ...

Write that down somewhere (other than only here). When the DocBase 
go public, there could be a place there for new ideas and suggestions.

[...] can't be done

kg$123 is not that bad - other languages have much weirder syntax.

maybe, we can do 123#kg but i don't see a big improvement here.

kg#123 may be possible too.

or 123$kg