World: r3wp

Geomol - my question was rhetorical. I think I do understand what 
Gabriele means, I just don't agree with the outcome. There are clear 
places where to post, easy as that. It is a bit difficult sometimes 
to get Carl's attention, but 80 tickets a month get such an attention. 
The development process of R3 might look chaotic, jumping from one 
area to the other, but if we want, and we care, we know how to get 
such an attention. 


I for one asked Carl privately about your concern towards R3 speed 
in certain situations. And you know what? I got some answer too. 
I asked Carl to comment to your ticket, he did so. In few hours. 
You could do just the same, no? It is very easy to become a naysayer, 
to express some worries, etc., but other thing is to actaully act, 
not just talk, and then your saying applies - "less noise and more 
thinking (and acting) would be good for a change" :-)


.... and please - I think I don't need any guides on what should 
I comment, or not. But the fact is, that I don't want to let anyone 
to dismiss the hard work which is being put into R3. I don't care 
about myself at all, but I see it at least as dishonest to those, 
who really try to bring R3 out, and we have few such friends here 
...

BrianH: 'call in R3 is more low level, right? I would like to do 
following thing - I want to have few powerpoint presentations running 
in a loop. In R2, I could use call/wait, and once the presentation 
is finished, the console returned. In R3, there are no refinemenst 
as /wait, /output.


Is there any equivalence? Was R2 using "start" command in windows? 
I tried with "start", but got following error:

>> call "start /WAIT powerpnt /S test.ppt"

** Access error: external process failed: "Systém nemùže nalézt uvedený 
soubor.^
M^/"
** Where: call
** Near: call "start /WAIT powerpnt /S test.ppt"

hmm, even if I use: 

>> call "cmd /c start /WAIT powerpnt /S test.ppt"


... it behaves in an async manner. It is a bit difficult without 
some R2 features. I had another plan in mind - I still have some 
R2 wrapper for Win32API funcs like launching app, moving window, 
setting its position, finding window by name, etc. (it originated 
from Gregg's send-keys dialect). But with R3, we don't have DLL interface 
anymore.


Now I wonder, if I should write an extension, or there is some plan 
to adress #1223?


I can use R2 as an interim solution, just wondering which way we 
go in R3, in regards to 'call. If I should start to think Extension 
way, or wait for 'call being enhanced?

you mean you're going to use REBOL TWO?  why?  I thought we should 
all be using R3   ;-D

I plan to use R3. I defined what makes R3 beta a good release, and 
adressing 'call is one of those points. CGI/IO was already adressed.

Well, I need to calm down ... I am out of this place for some time 
....

Pekr, I did this interface some time ago. It is based on a previous 
implementation from J.F. Theis, and is implemented as a TCP-IP server. 
I will certainely make out of it an extension for R3, but it can 
be already be used either with R2 or R3 as soon as you are able to 
send TCP-IP commands. See http://colineau.societeg.com/rebol/r2win151.html
. Unfortunately, it seems that the host site is down right now. If 
the shutdown persists, I will put it  in some other place.

Gabriele, every time you mention a document or code that was posted 
somewhere two years ago, without providing a link, or stating which 
AltME world the file is in (with file name/path preferably), then 
it comes across as useless complaining. If you want something done, 
say so. If you want to say that you *already did something*, prove 
it. Show me. Complaints about a time before the restart of the R3 
project aren't relevant to the current project. Live in the now :)

The standard TELLS you when

 - No it doesn't. The standard doesn't cover R3 internals, not even 
 in a generic non-language-specific way. The "when" I was talking 
 about has nothing to do with the encoding itself - it has to do with 
 internal data formats.

As for where things go, we built places for such things (as mentioned 
above) and they haven't moved in a year.

wouldn't it make sense for TRIM/WITH to be able to use a bitset?

>> trim/with "abc" charset "ac"
** Script error: trim does not allow bitset! for its str argument

Today I have seen a Wii GUI in action. It has been designed to be 
used with a remote controller.
Also XBOX 360 and PS3 have been.

Interactive Boxes like Digital TV receiver, Mediacenters are designe 
to be used with a remote.

We are entering in an era where mouse and keyboard are no more the 
standard input methods.

To the designers of REBOL3 GUI please consider the new paradigms 
and provide different interaction methods:
- GUI to be used with REMOTE controllers and similar devices

- GUI to be used with the click of the mouse an keyboards and even 
pedals.

- GUI to be used with multi-gesture multipoint touches (either on 
big and small screens)

Animated transitions and some 3D are necessary for a modern GUI system.


GUIs are the basic instruments users interact s with our applications. 
If we give the feeling of a modern GUI 50% of our work has been done 
because they will feel the program to be modern and good, even if 
it isn't. really so.


Our customers are people: specialist and families like the one I 
have encountered this evening. They use Modern Touch based Cell Phones, 
MediaCenters, Remote Controllers and at the and Mouse and Keyboards.

Hope my observations helps.

This has been a very silent month for REBOL... probably a lot of 
things are underway.

Carl is furiously at work putting time on the host.  a lot of unglorified 
but required time.

once that is out the bag... I expect a lot will suddenly happen at 
once.

its "the deep breath before the plunge"  ;-)

A95 released with compiler optimizations.

I see a performance increase. R3 is faster than R2 in a few tests, 
I've done so far.

I tested this:
dt [a: 1. b: 2. loop 10000000 [a + b * a / b]]
and this:
do http://www.fys.ku.dk/~niclasen/rebol/r3/mandelbrot.r

Also
do http://www.rebol.com/speed.r
show an increase in REBOL-Hertz.

I get with R3 lower performance with Processor, Memory and Disk/File, 
just the value for Console is almost 500% increased

sqlab, did you upgrade to latest alpha 95?

Here is slight comparison with the latest R3 release:


I used this script identically for all tests: http://cyphre.mysteria.cz/tests/mandelbrot-int.r

results on AMD Athlon 1.4GHz, 1GB RAM:


REBOL2 partially JIT compiled version	 	0.471s      1.0   speed ratio
REBOL2 (REBOL/View 2.7.6.3.1 14-Mar-2008)	12.15s    25.8 x slower
REBOL3 (r3-a95.exe)					13.87s    29.45 x slower
REBOL3 (r3-a94.exe)					17.54s    37.24 x slower

yes, 
 Version:   2.100.95.3.1
 Build:     24-Nov-2009/6:42:02

msvcrtd.dll disabled

Hm, I need to test some more, I guess, because I initially see a 
speed increase, but your results show differently.

Same mandelbrot:
REBOL 2 ver. 2.7.6.3.1	9.35s
r3-a95.exe			7.521s


Results is from an Intel Pentium 4 CPU 2.80GHz running WinXP. Maybe 
R3 is compile optimized for intel CPUs?

sqlab, how do I check, if msvcrtd.dll is enabled/disabled?

just rename it or put it somewhere out of the search paths.
It was needed for the first a95 release with debug information.

Ok, got it. More info here:
http://www.rebol.net/cgi-bin/r3blog.r?view=0297#comments

Gabriele, is it possible to dispatch multiple request to wiki TCP 
examples "pong" server listening on single port? It should be possible 
but for me second request is without response until the first still 
open. Your HTTP scheme is too much complicated to me as lecture reading 
:). I've tried to transform rebol.org webserver to R3, I've got response, 
but seems to me useles to serve one and one only connection at time 
when the port is asynchronous by nature. Any hint?

We've been through this, but I still have to bring it up. "REBOL 
3 needs a new name"

A new name, a new beginning, and a new reputation.

What was the consensus last time? I still disagree about changing 
any names, btw.

I decided to use REBOL to develop our system a couple of months ago, 
and everyone was against me. Now the system was done, they were surprised 
at how quickly I did this using a scripting language. They considered 
this development process a good example of the productivity of scripting 
language. They want me to talk about it in front of 400+ employees 
in the tech department. But I am too afraid to say REBOL out loud 
in front of people now. I will call REBOL "a dynamic language" in 
the tech convention. I just cannot call it REBOL. Last time I did 
this, everyone was question me.

So please please please rename REBOL to something else.

I remember Carl like the name "AltScript"

REBOL 3 beta is coming out soon. If we want REBOL to have a new name. 
This is the right time.

I would be also in favor of  renaming Rebol. The impact on those 
who already know or use Rebol would not be negative, and it could 
be a good opportunity to launch a "new Rebol ", with a new image 
to the other.

Jerry, make sure you mention what REBOL stands for:

Relative Expression Based Object Language

I'm against changing the name for several reasons:


- The name is established. It takes years to re-establish a new name 
to gain new momentum. Carl's company name needs to change as well.

- You need to relate REBOL 2 with <new-language-name>. This requires 
effort and you'll have to waste a long time explaining this fact.

- There are already too many *-script languages out there, which 
makes AltScript less distinctive. So, is it a Javascript or Applescript 
derivative or what?

Downsides to REBOL:


- The name is overused, which is why there are people against the 
name.
- Pronounciation problem.
- Pointy-haired boss problem (REBOL? REBEL? Hackertool.)
- REBOL is sometimes confused as a COBOL or ALGOL derivative.

- Had REBOL been named AltScript in the beginning, there may not 
have been a problem, but also no distinction.

Suggestions:


- Tone down the use of the name. Use the name as for a technology 
engine, rather than something end-users will confront, like DirectX 
or OpenGL.

Jerry, why do you think that people who have problems with REBOL 
name will not have same problems with any other name? I think that 
changing name will not make things easier. They will be scared in 
the same way. Mostly that   using something new is hazardous. For 
example if you are the only one in the company who knows how to program 
in REBOL, what if you will have some incident suddenly? This is nothing 
what renaming can solve. Anyway, such a discussions are usualy in 
Advocacy chat.

I wouldn't change it. Python, Scheme, Java not very elegant as well.

I agree with Henrik, both on not changing it and on the problem list. 
 I've gotten past all that at work though, the development group 
knows what REBOL is and can talk about it in conversation just like 
they do Ruby (they are Java based primarily).  One thing I do find 
that seems to give some relief on the name issue is just to call 
it R3.

I also agree with Henrik, and believe that you really need to stress 
that the name was derived from an acronym as pointed out by Henrik. 
 Once that is fully understood, the relationship to rebel or the 
negative meaning should be a thing of the past.  The benefit is more 
people will be aware of the acronymn that was used to derive the 
name.

I have colleagues that called Rebol  REBOLa, likening it to a virus 
after a colleague's machine acted up after installing Rebol.  It 
was coincidental as I have never in my years heard anyone else complain 
of a similar happening.

Carl should also emphasis the name being derived from the acronymn, 
directly on his website in big bold type, so it is more understood.

AltScript sounds like some yet another scripty thing to me.. nothing 
to remember it by .. They are scared because they don't know it, 
probably never heard of it so it's that what we have to change. I 
think the R3 openeness and new features will make it more approcable 
by the masses so there is a chance to spread. Also rebol community 
is qute self sufficient and closed in it's own world which doesn't 
help at being "everywhere".

..

If you want it to sound right to the enterprisey people you should 
rename it to **JavaLisp** now that would be something that sounds 
right :)