World: r3wp

Thanks for all your work on World, and the QA John. It's very exciting.


As far as the outrageous bytecode size, from one who hates bloat 
as much as the rest of you here, I am happy to make certain tradeoffs 
in return for other benefits. If I have the choice of REBOL, World, 
Red, and Topaz as "mindset compatible" languages, I can choose which 
one to use for a given purpose. Yes, that opens up other questions 
and issues, but I'm happy to see more REBOL-like langauges.

John, on range! values, do you envision transparent iteration/lazy 
generation support for them? I use the term 'bounds for what you 
call range! and my 'range is a series with the values filled in (with 
/skip support).

I'd really like to see such a combination, a fully working glass 
with a "compiled" rebol clone .. I just have too many performance 
issues with what we have at present.  It may be fast to write .... 
but ...

Thaks for the kind words, Gregg. I was very much in doubt when growing 
the instruction size to 256 bit, but I must do something right, as 
the performance shows. This is an alpha, and things will change. 
And I haven't done much compiler optimization.

*Thanks*

About range!, I haven't thought of what you describe. We can talk 
more about it, when you're able to try World.

btw, I really like the   call-by-word   idea you show in countdown: 
 2

I think the range datatype should be tweaked a little:
 block/4-3
should return.
== [d c]
 

otherwise  it becomes mightily confusing, since as you have it now, 
when going backwards, its 0 based and going forwards its 1 based.


also,   /4-4 and /4-3 shoudn't return blocks of the same length, 
its almost impossible to compute ranges since we'll have to add IF 
to know if the start and end are the same, and then decrement them 
by 2 instead of one, for example... pretty odd.

I'm not sure about that either Max. We'll let John catch his breath 
and he may have a reason.

World alpha has been released.

Who's the first one to successfully run a test?

Well, I ran the test and just created a function. I love the automatic 
quotes by the way.

I just renamed to exe, tried to run from Total Commander, and got 
the following error:

The NTVDM CPU has encountered an illegal instruction ....

Ok. Others got the same error? Anyone has an idea why?

Test
m

Both loook good!

Re: naming convention from ann-reply

You could use the same names as they are now: world_linux.xxx and 
then suggest the user can/should rename his version as world.xxx. 
That way a person can get going immediately without that step if 
he wants to.

Ok, suggestion noted. I'll think about it and maybe change it later.

World should accept REBOL [] as header to run R2, R3 scripts through 
it without editing these.

First let's see, how many REBOL scripts will actually run. I assume, 
a lot of work is needed in %rebol.w . You guys are welcome to contribute. 
Maybe we need a project for that!?

Pekr, maybe this will help: http://support.microsoft.com/default.aspx?scid=kb;en-us;314106
with that error.

Geomol, interesting. I can execute the mandelbrot.w but not the cmandelbrot.w 
w> cmandelbrot
** Error: cmandelbrot has no value
** Near: cmandelbrot

w> do %test/cmandelbrot.w

I don't think, there is a cmandelbrot function.

My bad. I was thrown off by the way the original "test" worked.

I thought you could simply type the name of the file. :-)

:) Look at the bottom of the %cortex.w file. You'll see the test 
function and the m function. I just include them for now, because 
I call those two scripts all the time.

To have more luck with running REBOL script with world, do the %rebol.w 
first.

%rebol.w will extend World with some REBOL stuff. We could say, REBOL 
then is a dialect of World.

I added cortex_alpha.pdf to World at GitHub. It's some initial documentation.

there's something with mime-types imo - right clicking the exe, choosing 
to save, saves it as world_win32.exe.htm, containing github html 
stuff ...

Why you just down download https://github.com/Geomol/World/zipball/master

Because thing shoul be easy! And if Geomol points us to the link 
he posted, I will not look around to find some mysterious zipball 
...

so try to click on it using left button:) https://github.com/Geomol/World/blob/master/world_win32.exe

OK, I found I still have Tortoise git installed, and synced the stuff 
....

It is quite a long time I last used it to get R3 sources ;-)

It would be better to contact github to add colorizer for the World 
language.

Clicking left button does not help at all - brings you to subpage, 
where no download button is available. So I used raw link. It stores 
it as 515kb exe, which is not runnable.

It would probably require a small fix in the current colorizer used 
for Rebol (so World is recognized as well)

of course there is the download link... it's called RAW.. and you 
could also see history with diffewrent versions

(once there will be a new versions)

Just try to download .raw - it is not runnable

OK, now the question - World is some 515 KB - it is still small in 
comparison to many other tools out there,but still the question remains 
- why it is so "big", as it mostly compares to /Core? :-)

Interesting - after the start, it also takes 4.3MB vs 2.3MB of R3/View

It would be intesting to have 'stats (profiler) available - anything 
like that planned?

Size is probably because of some Windows includes/libs. From my mac:

-rwxrwxr-x+  1 john  staff  360286  4 Dec 09:53 world_linux
-rwxr-xr-x   1 john  staff  346216  3 Dec 11:27 world_macosx
-rw-r--r--+  1 john  staff  515991  4 Dec 09:50 world_win32.exe


The OS X and Linux version is much smaller. I have to dig into that 
at some point. But I don't think, it's so important right now.

Geomol - absolutly - not impotant, just a question of couriosity 
.... You better focus to some other areas :-)

Yeah, STATS would be nice. It's not high on my list atm.

I have got first crash. Try code like:

for i 1 1000000 1 [print "ahoy"]

... and press ctrl c

ctrl c terminates World. That's intentional.

ok

I guess you should give us a bug tracker.