World: r3wp

mine is here - http://curecode.org/rebol3/ticket.rsp?id=1223&cursor=26

So far, R3 for Linux is only there for 32 bits x86

If you're willing to test the host kit, I think Carl will be willing 
to bump the priority for other architectures

CALL having /WAIT hardwired is better than the reverse. Without /WAIT 
it's useless in most cases

You'll have to upgrade to Linux, Petr ;-)

... or stay with R2 :-)

now now pekr... I'll hold you to it... as you said before, everyone 
should use R3...  ;-)

Ah, calls to the standard C system() function are in src/os/posix/host-lib.c

Maxim, I didn't want to remind Petr :-)

hehehe  I keep reminding... I just like to rub his face in it everytime 
he mentions R2  ;-)

Very good :-)

OS_Create_Process simply does

system(call);

OS_Browse does

system(cmd)

with a constructed URL

Both have an extra

int reserved

parameter, so presumably that would have to contain a code for the 
REBOL refinement

It's not clear how you could use that to implement complex refinements 
such as /output

its possible Carl left some things for further consideration... we 
did pressure him into releasing sooner than later.

call refinements would sit in well with something easily left for 
later.

Well, quite a bit of extra consideration is needed here. There are 
no error codes returned, either

Problem with the parameters is that the host interface will change 
when added later

Maxim - you are off once again ...

Call has nasty bugs, which are in CC for quite some time. It is just 
general disagreement, on how Call should be implemented. BrianH stated, 
that it is now more low-level, and I objected, that I don't care, 
or it is not usefull for most use cases ...

Maxim is right that they are not bugs, but a very primitive implementation

I am now surprised the implementation differs between the platforms. 
Call is one of necessary beta quality fixes ....

ah ... well, how is that it worked with R2?

Carl once said that the C code for R2 CALL is 1200 lines. If he wasn't 
joking, he has saved 600-fold over that so far

BrianH is probably right, that because the code is part of host, 
we can fix it ourselves

1200 lines? uh :-)

it had  some quirks and evolved at almost every iteration of the 
language.

Yes, I expect you to start coding now :-)

Petr, that is

I don't care for coding, I prefer to complain around, until it gets 
fixed ...

We know...

Kaj - so stop using the wrong attitude ... you just fit old R2 mental 
model :-)

Problem is, it's just complaining. You think it makes things go faster, 
but it doesn't

I just analysed R3 code for you. How does that earn me an accusation 
of R2 mindset?

http://www.rebol.net/wiki/REBOL_adoption#.22You_can_code_it_in_few_lines.22_attitude

Well, here's the news: by releasing host code, Carl has just extended 
that principle to the C code, as well

You see? I don't care. I can see it as an repetitive pattern, which 
is very wrong ... 80% of potential newcomers will NEVER actually 
care to implement anything themselves. I have seen the attitude with 
Gabriele, Holger, RT ... I hate it

That would be excellent, because 20% of your users contributing would 
be phenomenal

Not everyone can code. But I proposed bounties. I sponsored 500USD 
for networking protocols ...

Word just came out that even the VLC project, one ot the most popular 
programs ever, doesn't even have any contributors left for the Mac 
version

Yes, that is sad (VLC)

I still think, that Amiga/MorphOS like bounties could be motivating 
factor for some ppl to pick-up and deliver some stuff for R3.

pekr... one problem today is that people have come to an assumption 
that software programs itself.    I don't know where that is coming 
from, but in ALL real projects... 


at some point, you actually do have to raise your sleeves and put 
a little effort into it.

We actually find it uplifting, because it means things aren't as 
bad as it may seem for Syllable