World: r3wp

For FBSL it was a false positive also.

yeah ... I wouldn't use AVG for that reason too.

Turns me right off ..and also comparisons that show AVG is the worst 
perfomer of the free antivirus software out there.

Avira marked the same file as bad. It was a false positive for another 
reason - are people already writing malware in R3?

Every once in a while these antivirus softwares mark some innocuous 
component that happens to be used in malware as bad :(

The last false positive I got was the Windows port of BSD's firewall.

There's a lot of new code in there!
   alpha 30 thru alpha 36 -- 530K
   alpha 37 -- 531K
   alpha 40 -- 533K
   alpha 41 -- 568K
   alpha 42 -- 576K

We can now load JPGs, PNGs, GIFs and BMPs.  Only adding 46k is pretty 
amazing.

we've got 100th user to ugly console rebdev chat :-) so - we are 
growing :-)

Ammon - it seems we can't save in some formats, e.g. GIF ....

save : not gif. Only PNG, BMP like R2

what about the gui upgrade & performance  with table ?

not there yet. still working on getting the basics straight.

Don't treat Henrik, he currently has some problems with SET functionalyties 
:-)

It was R2 related, actually. :-)

no problem.......... l  wanted to know ..........thank for your effort

you can still follow my screenshots at http://rebol.hmkdesign.dk/files/r3/gui/

yes i check it often ;-)

The process now is to clean up the skin, making it as simple as possible 
first (no gradients or prettiness for now), so Carl will be able 
to approve it and so I've started from scratch, fetching in the styles 
one at a time. During this process, I've exposed a few problems and 
solved those, so it was a good idea to do that.

Just a question: will REBOL3 have more Obejct Oriented capabilities 
than REBOL2 ?

You can use path (index) with objects in R3, like you can with blocks 
in R2. Like:
o: make object! [a: 1 b: 2]
o/1: 42

like what ?

geo, are you implying that path notation for blocks has been removed 
from R3?

But if you think in lines of classes, private/public content etc., 
then I don't think so.

Maxim, no no, you can just also do it with objects.

private/public ??? who need that

phew!

steeve are you being ironic?

There are ways to make private data with object in R2 aswell, using 
BIND, if I remember.

yes you can bind code to two different objects, its quite easy in 
fact.

but you can't bind properties of one object into another one

define properties

no inheritance either

you can have a function in one object which executes code from anothe 
object using its own values.

attributes/vars of objects

What I meant is:

o: context [
	a++: show-a: none
	use [a] [
		a: 0
		a++: does [a: a + 1]
		show-a: does [a]
	]
]


Whis way a is hidden, and you have two methods, one to add 1 to a, 
and one to show a.

I prefer that

context [
	a: 0
	set 'o context [
		a++: does [a: a +1]
		show-a: does [a]	
	] 
]

Nice one!

i use it all the time

And I use that a lot myself too.

Almost all my programs start with:

REBOL [ ...]

context [
...

so, we don't need that private/public stuff

It is planned to get access controls for object! fields, but we don't 
know how yet (haven't decided).

REBOL 3 already has module encapsulation (as of last week).

Also in modules.. once there will be basic functionality (if I understood 
well)

We also have multiple inheritance (kind of):

>> o1: context [a: 1]
>> o2: context [b: 2]
>> o3: make make o1 o2 [c: 3]
>> o3/a
== 1
>> o3/b
== 2
>> o3/c
== 3


I can't remember, what else is in OO theory. But don't we have most?

its not dynamic inheritance

It's not. We do dynamic inheritance manually in REBOL. Our objects 
are object-oriented in a different way than class-based langs.

but having classes is MUCH faster. I mean exponentially.  as in liquid 
we are talking about 100 times on large datasets.

and with huge datasets it becomes wholly impossible to work with 
rebol's object system.