World: r3wp

R2 does not like it when the block under execution gets reallocated. 
I think that the same was true for PARSE. I don't remember about 
FOREACH etc.

Nenad, this group is web-public, so...

I don't know if you can really make something nasty happen out of 
this, though it cannot be excluded of course.

Gab, yes, I checked after posting...anyway I guess that people following 
AltMe channels are here, not sure there's much others following only 
on web and not having a account here.

same could be said for RAMBO :-)

I thought that RAMBO tickets wouldn't be indexed by search engines...after 
testing, it was a wrong assumption.

Anyway, I'm in favor of exposing flaws as long as they can be fixed 
by someone.

isn't this conversation indexed by search engines?

(i'm not saying it should be posted to rambo, i'm saying the cat 
may be out of the bag already. maybe a direct message to Carl would 
be a good idea, though i wouldn't hold my breath about R2 updates)

I think that Sunanda could be able to filter out Geomol's post and 
my reply from the web export.

I've notified Sunanda about that.

IIRC, BrianH is something like R2 release guru, we should ask him, 
maybe he has some special connection to Carl :-)

Or maybe its Carl's brain simulation is good enough to fix it. ;-)

ladyslave the main chief guru referes not to you in my mind i don't 
even knows who you damn are ...

the main chief guru in my mind it's carl the guy with all the answers 
and all the keys and keeping them for him ...

how do you want people to be more implicated in rebol VM  enhancement 
if  they can't learn from what already exists in it ... and once 
again i'm sorry but it's a mater of fact compared to public accessible 
open source VM  REbol is clandestine...  and what is the purpose 
to do a revolution that is kept in a bottle ...

ruby, perl, tcl/tk python are all doing now in days computing history 
where is rebol in this picture ?

and the problem is not R2  or R3  or R20 ... you see ... the matter 
of fact carl tomorow decide to stop rebol we are all fucked up ... 
Openning the source code of rebol is a way too write it in history 
many where the langage with close  source VM. No one now in days 
remember them and the only ones  getting all the attentions are the 
ones with GPL  like system

Shadwolf - answer just one question for me - how open-sourcing R3 
today helps to finish us the product? It does not. So please - save 
your open-source rants for future, once we are stable, once R3 is 
finally out there, then please ask Carl to reconsider the situation, 
maybe he will be willing to open-source REBOL, I don't know ...

Pekr open source or not open source is a long rate debat man it's 
been on the table since day 1 I met rebol 6 years ago and what i 
can see is that the rebol community is skrinking and not extending.

The trouble is, that you constantly repeat your point of view on 
all possible places, whereas we have more important things to do 
right now. REBOL should be open-sourced long time ago, or we should 
wait a bit for open-sourcing it in future. And open-sourcing host 
code is good for starters. Yes, we are slow, but we are getting there. 
Carl needs to finish Parse, then he is back to Extensions, which 
apparently are going to be used for Host code isolation. Once that 
is done, the code can be released.

Open sourcing product, which is still in alpha, undergoing structural 
changes, is not a good idea imo ...

Yes, REBOL community got shrinked. Because our only product is R2. 
And R2 is stagnating .... because most resources are now dedicated 
to R3, which is not production ready yet. Once in beta, or 3.0, we 
can start looking for new users, and we can start to reconsider to 
ask Carl to open-source Core. I see no point raising such questions 
nowadays. Let's wait half a year at least ....

Pekr for example by  solving directly the bugs instead of filling 
a bug track and than wait for example ... you still can document 
on a bug track what evolution you maid... The question is can carl 
work on all the topic related to rebol at the same time ? the answer 
is no. And until now carl take a topic enhance it correct the bug 
related to the new topic addition and then move on to the next topic 
.... i'm not sure that's the more efficient way

Shadwolf - you are most interested in View, no? All even system, 
all devices - file io, tcp, events, etc., everything View related, 
is going to be part of host code = open-sourced.

Shadwolf - what are you talking about? Just look at CureCode. CureCode 
is absolutly cool and it much helped to shape-up R3. We get 100 tickets 
implemented a month. That might be even more, than you might get 
with some even open-source systems. And you also make it sound, like 
there is no plan, whereas there is a concrete plan for beta - http://www.rebol.com/r3/project-plans.html
.... Carl even offered to wikify it, so that the community CAN influence 
release priorities ....

Pekr ... mainly cause i'm an idiot that doesn't understand how works 
parse ( in advanced mode.) that's the point you wanted me to go ??

Wrong i'm not interrested only in View or draw i made lot of kind 
of programs in rebol...

I think I have nothing more to say right now. Those who want to see 
the light at the end of the tunnel, start to see it, those who don't, 
- I am sorry for them :-)

I had no intention to go into Parse. My only note of Parse was, that 
it is the actually undergoing priority, and you can note the fact 
by looking into R3 Blog topics .....

...some people would love to make sure that there will be no more 
tunnels...

In fact - I am confused about what you actually want? What do you 
think Carl should work right now on?

Gabriele - if you find yourself in a tunnel (which is our situation), 
you first have to escape it ... which is what we try to do ...

Apart from that Shadwolf forgot, that it was not only Carl himself 
working on R3. It was also Ladislav, Cyphre, you (Gabriele), BrianH, 
although not all of you contributed at C level ...

ofcourse it is a priority to polish parse ... but the thing is only 
car can try new feature...  having opensource accessible to all mean 
alot of tries can be done and shown and that's better than the 1 
to  X  relation we have. Mainly and i'm not the only one to say it 
we propose alot of things but since we can't show them they are. 
what do i want is a rebol that evolve faster ... it's been 2 years 
since rebol 3  enhancement started we lost alot of time in futil 
consideration like how do we name that and how about changing the 
name of this ... and we skipped during a loooooooooong time the main 
topic ....

R3 initially was just a R2  version with UTF8  support ...

and then carl said ok lets redo it all and 2 years and around 100 
 alpha version havec been made and we are still in the process ... 
 Mainly why because it's the work of a single man... and that's all. 
If carl don't push the rebol car the car stop and that's all ... 
you see all the comments you can do or not will not have a real impact 
on how the car rebol progress or not.

Shadwolf - then please go and ask Carl privately about open-sourcing 
topic. It sounds like I am against it - I am not. I just don't think, 
in comparison to you, that open-sourcing it NOW would bring us R3 
any faster. Soon enough (which I ufff, repeat for so long) we get 
EXtensions and host code. I bet it comes in few months. Then ppl 
can start porting efforts and extending the language .... ask Maxim 
for his plans :-)

Shadwolf - R2 to R3 is not Python 2 to Python3. It is a complete 
architecture change ....

Shadwolf - I talked to Carl, and he wants go into beta Autumn 2009. 
I actually think it will be Winter 2009, but I would like beta to 
include all important Core stuff, including tasking ...

cyphre working on R3 actually ? Cyphre didn't loged to atme since 
july 27 how active is that ?  I don't see many topic poped around 
at same time i see 1 main topic wroked and the related things worked 
at same time that's all. 

Complete architecture change .... hum ? but that's came along the 
flow my brother it wasn't a from the begining statement.

Cyphre is not working on R3 AFAIK.

I am glad that R3 was main rewrite of the REBOL. R3 starts to shape-up 
nicely. It is REBOL we should get from the very beginning in 1997 
though :-)

did i said R3 wasn't good ? i like many features in it but the progressing 
is slow that's all and repose entierely on carl...

well, the ports system has been done for over a year and is documented, 
but where are the protocols?

but yes I would say R2 lasted too long ... carl was glued in endlessly 
bug correction and few where the real improvements... the major addition 
in R2  was the AGG/draw dialect which since then have not evolved 
a bit... look at the number of tickets in old rambo. So yes starting 
with a ew base and bring alot of enhancement was a certain thing 
the problem is along the process we lost alot of people mainly interrested 
in rebol but not seeing it evolving fast anough and that how do you 
solve it ?

Henrik - we are bunch of lazy and uncapable lamers :-D

I would like to know, if some good C coder having experience with 
cooperative development could tell us, how much faster would R3 be 
developed, if more Core ppl would be involved. You know things like 
coding style, etc. I wonder if it is even possible that multiple 
coders work on such kind of project, like language kernel? There 
must be high standards, no?

yeah... but we assume it :P or at least i assume my lazyness and 
my stupidity but in other hands most of things I done with rebol 
where totally unexpected things .. and  once again every time i faced 
limitations that forbided me to push them till the end and release 
and maintain a product more than a demo ...

Shadwolf - we do wish, and we do hope for those ppl to return. We 
also hope to attract many new ppl.

Shadwolf - what imo Henrik tried to point out is, that R3 networking 
seems to be kind of solid for 1-2 years, yet noone picked-up and 
rewrote networking protocols - and that layer is of cource open-sourced 
...