World: r3wp

Geomol - e.g. right now we are jumping to yet another topic - binary 
conversions. I think you might want to say something here :-)

That's fair enough. To make a bug report requires a substantial allocation 
of time and concentration. You have to check that it really is a 
bug and not a misunderstanding, that it isn't already reported, which 
requires searching through the database, and actually writing the 
bug ticket.

You also have to check that you have the latest version etc..

Oh, no, I already wrote the bug. :-)
Yes, it's a lot of work to report bugs.

New Ticket created with ID # 698

What frustrated me during all of R2 development was finding a bug 
that really threw a spanner into the works of some application I 
was making. I would report the bug, but it wouldn't be fixed for 
a long time, basically forcing me to abandon the application, or 
workaround in an extremely time-consuming and ultimately fragile 
way. What I would have liked was access to Rebol source code so I 
could just find the bug and fix it. The amount of work I put into 
workarounds... gee whiz.

For R3, I haven't gone to the same effort. I just don't want to go 
down that road again.

Having said that, it appears to me like R3 is steadily getting on 
its feet. If the linux rebol console can get some more user friendly 
features like arrow key history etc., for example, then that would 
make things easier. Only recently the scroll buffer was introduced, 
thankfully.

remember this group is web-public.

Good.

I really like what Maarten said.

Paul, I'm ok with this being public. It should be known, how different 
REBOL programmers should see the situation. New folks need to know, 
what they go into.

Ok, but don't complain when people don't embrace REBOL because they 
read comments like these on the internet.

Paul, those comments are everywhere, even on Carl's own blog, so 
I wouldn't bother concealing the state of R2 bugs.

I attracted Bobik to R3 once again. He was waiting, and waiting, 
and then he went into pissed off mode. So I told him to concentrate 
upon R2 (he is now using RebGUI) and Python, and just take R3 as 
- "might be" a good thing in the future. We now have good chat about 
the progress, and there IS a progress, although it might seem to 
be slower than we might wish for ...

I hope, I'm not complaining too much. I see my comments here as just 
stating, how I see the situation. How long, I feel, it'll take for 
R3 to be, as I would like to see it. It's just my opinions. I would 
love to see e.g. my paint program on all platforms. I realize, I 
don't get that using REBOL. So I have to figure out, what to do then.

R3 is whole new architecture, and in few years, it is goint to be 
cool. As Maarten said - maybe some completly new stuff appears, which 
blows us all. But so far, I still can see the chance for R3. There 
is not much change with Python, Perl, Ruby, Lisp, Scheme, PHP, whatever 
- no revolution happening - each technology has its place. There 
is also not revolution with the web aproach. So, just a patiecne. 
And those impatient - forget R3 for one year, then come back and 
investigate the situation, to stop your frustration ...

You should also note that for ppl who do care, and are trying to 
help, such comments are really frustrating.

Why do you see it as frustrating? I it because I tell the truth? 
I hope, people dont live in a fantasy world and wait and wait for 
years.

I've been programming C code more and more. Rebol is implemented 
using C, and extending Rebol with C gives it great new powers.

Geomol - on one hand you claim, that you don't have even free time 
to report a bug, not to mention to help with R3, while on other hand 
you claim, that you look for a platform/language to port your app 
too (which surely is huge effort)? This sounds like a discrepancy 
to me ...

Geomol - ppl who do nothing for R3 do wait. Others try to help here 
or there, as their time permits ....

Pekr, I've tried to help for a long time in r3-alpha. For years. 
I've had my fingers into C source for Windows, trying to figure out, 
what's going on. I've just come to a point.

Geomol - anyway - if we talk about R3 and open-source, then remember 
that only host-parts plus View are going to be open-sourced. The 
language itself (which is the culprit of bugs you are pointing out) 
is going to be closed source anyway. So the only thing we can hope 
for is, that at least not caring about host code and View Carl will 
have more time to spend on the kernel.

Yes, I'm aware of that.

But - one thing worries me anyway - R3 http scheme is out for how 
long - 2 years? And we don't even have proxy for it. Not to mention 
other protocols. So if we (community) can't find some time to do 
other schemes, where's our help then? The thing is, that schemes 
stuff is really difficult. I can imagine myself doing some easy styles, 
but surely not complex thing as network protocols :-(

Who did the original schemes? Holger?

Pekr, this example with proxy, we see things differently. As I read 
your words, you're complaining, we don't have proxy for the http 
scheme (after maybe 2 years). I can only talk for myself, but the 
reason, why I don't look into things like that, is because it's way 
down on Carl's list (maybe third or fourth priority). I would like 
to see the first levels bug-free and stable, before I move on. (And 
you don't hear me complain about proxy. :-) )

N

Geomol - new http 1.1 scheme was done by Gabriele. And the only reason 
I mentioned it is, that it is an example of open-sourced stuff, where 
community could help, yet nothing like that is happening. Now what 
does it have to do with how deep is such stuff on Carl's list? - 
it is open sourced. And once again - see your arguments. So you DON'T 
know there are any networking related bugs, yet you claim you are 
not willing to work on it, as you suppose there are some "first levels" 
related bugs :-)

Anyway - no point for me to participate in that discussion. I stated 
my opinion clearly - ppl either do want to help, or they don't. Choose 
your side ...

Rubbish, Pekr! :-)

I wanted to help. But I won't sit on 123 floor in a building with 
unstable basement. How should I suppose to test my work? Think! :-)

First you crawl, then you walk, then you fly.

I will sit on a 123 floor with unstable basement because the view 
is beautyfull  ^^

I wish, people wont see my words as complains. I'm not complaining. 
I'm just stating the facts and how I see the future of the R3 project. 
I would love to be proven wrong. In the meanwhile, I will have to 
figure out, what to do with some of my developments. I love R2, and 
will continue to use it daily, because there's nothing better out 
there for many things. But not all things.

shadwolf! :-) I love that! And I'm also a person, that like to live 
my life dangerous. But I'm also practical in many ways.

R3 is a work in progress it's normal to not already do heavy things 
with it. But at same time not using it doesn't help to find the problems 
it has to solve them

I used it as many others. I and others found problems on lower levels, 
reported them, but some are still not solved.


Take money! as example. At some point, it became impossible to specify 
currency with the money! datatype. When did that happen? Why wasn't 
it noticed (maybe it was, I really don't know)? Why hasn't it been 
fixed? Other bugs depends on this datatype. I won't do the detective 
work to figure this out. Proper testing sessions would have found 
it, as soon as it emerged.

I just realized, it's been pointed out as ticket # 576 dated 28-Jan-2009. 
It has status "Waiting". For what?

It's a bug in the first of Carl's list of alpha testing priorities. 
We should be far beyond this level, but we're not.

Waiting: See the first comment.

So it's not known, if it's intentional. Did Carl write anything about 
money and currency? If not, then there's nothing to discuss. It's 
an emerging bug, as I see it.

(these things will always happen as long as the development process 
is not open.)

Gabriele, would you like it to be open? How do you see the R3 project 
heading?

Gee, how is money datatype really important? I NEVER used such obscure 
datatype even in R2. How is that a show-stopper for you?

Gabriele - I am not sure it is about stuff being open sourced. Even 
open-sourced development needs some organisation. Carl is organising 
few ppl around him, who are willing to work on some things. Protocols 
are open-sourced, RebGUI is open sourced, yet such stuff waits for 
real gurus to work on them - both to improve and to fix them.

Money as a datatype is useful as it is implemented as a form of BCD 
-- hence more digits of precision:
http://www.rebol.net/r3blogs/0094.html

Money! datatype is required for business apps that need exact counting. 
Just try to write an application doing simple maths operations on 
dollars or euros and you'll  see that decimal! is not an option.

Sure - it is a known thing, but I would not call it being a show 
stopper. We haven't had this feature for 13 years, so why should 
it be important right now? And if there is a bug, just claim your 
interest and it might get fixed.

Talking about that : http://www.rebol.net/cgi-bin/rambo.r?id=4363
This one was a show-stopper for me recently. Seeing such bug scared 
me about using money! datatype in R2. I had to rewrote several parts 
of a customer product using integer!-based fixed precision instead, 
just to be sure that amounts in the product would be correct.

For me is R3's show-stopper missing hash! and still buggy map! which 
is supposed to replace hash! So I agree with Geomol, that the core 
bugs should be fixed before playing with the high level or new functionalities 
(like threads). But I'm not so pessimistic, the things move on. And 
I don't need R3 so much now. I can do what I do in R2.

It's clear, that most series related bugs are because of unfinished 
unicode implementation.