World: r4wp

Graham: I was thinking it could be ready for end of year of beginning 
of next year. Currently Red is not for end users, and won't be for 
a while. Red/System could be used in production though, but I don't 
have time to write users docs or tutorials for it.

of beginning
 => "or beginning"

Don't we need to wait a bit for more functionality from red before 
writing user docs?

No, you can take REBOL as a model, all basic features will be identical.

We could then upgrade such docs later with Red's specific additions.

Who implemented Try REBOL? Was it Kaj?

http://tryrebol.esperconsultancy.nl/

So - we have something to start with, no?

I'll most probably add Red to TryREBOL. I need to integrate it in 
my CMS, anyway

The next step would than be to get the REPL for Red.

This is to try more Red beyond the things that are save to allow 
on a webserver/browser.

Is it necessary for a tryred to have Red be able to process cgi on 
the webserver.

The easiest way would be to use Topaz and "just" add a Red-to-Topaz 
converter.

That sounds like a lot of work. On TryREBOL, Red and Red/System will 
be used directly

Agreed. Also a great introduction like http://learnyousomeerlang.com/
would help using the REPL and compiling Red.

Kaj btw length-of is scaring me ;)

Is English scaring you?

There are OS- prefixes left in all platform files except Windows

I meant the discussion about the -of postfixes. :-)

Kaj: fixed.

More scary discussion of -of: http://issue.cc/r3/1818:)

Red is now Unicode from end to end: http://t.co/FR8vNV65

Cool!

And with string auto-scaling from 1 to 4 bytes per codepoints. ;-)

*codepoint

I will publish the new code tomorrow.

UCS1 to UCS4 scaling? Fixed-length internally?

Yes, Latin-1 / UCS-2 / UCS-4

Conversion for printing in UTF-16 done on-the-fly (no additional 
buffer needed)

Brian: you don't read Red's blog? :-) See http://www.red-lang.org/2012/09/plan-for-unicode-support.html

Been busy :(

Will you eventually be doing the same trick R3 does of keeping its 
symbols in UTF-8 format internally, for binary hashing? Of course 
you might be handling symbols completely differently...

The symbol being the string part of a word.

Yes, I currently keep an UTF-8 version in cache for each small string, 
but I'm not sure I will keep it.

I'm not using currently hashing for words, but indexing as lookups 
are done mostly during compilation.

I'm talking about eventually, when/if you get real runtime word types 
:)

Words dynamically created might need hashing though.

Right, it could be a good option.

I think that R3 stores symbols in something like a BTree.

Red stores them in a block! value currently, so probably the less 
efficient way...we'll see in the future how to optimize it the best 
way.

Maybe a Trie would be the best data structure for symbols (I have 
used them in other projects for fast short strings lookups).

One gotcha we ran into with R3: R3 uses aliases to handle case-insensitive 
word equality. We found that aliases were a good, safe use for that, 
but not safe or stable to use for anything else. Perhaps you might 
consider this for Red.

Maybe, I haven't decided yet how I will handle that.

I can think of using to-head and to-tail. My opinion is you should 
have a more descriptive name for your variable. So length-of-cityname 
instead of length. The length could be of anything and the variable 
could be reused. Besides native English speaking programmers have 
always faced the fact that their prime candidate for variable names 
were taken because they were a reserved word. Imho so is should stay 
length? and not be length-of though that last one saves me from pressing 
a shift-key.

Maybe it's too soon to think about optimizations, but what about 
function inlining?

(sorry that my question is out of topic... I was offline when I send 
it without noticing the unicode news:)

Optimizations are for Red/System v2 (the rewrite in Red). The compiler 
will have a different architecture much more suitable for optimizations 
(including function inlining).

Doc, how will the open sourcing of R3 have an impact on your Red 
work?  Will you be studying RT's code to see how everything was accomplished?

well, maybe depends upon the licence Carl selects? The other thing 
is - Doc might study some internals, but still he can't directly 
use C code, as Red is being written in Red/System, so C to Red/System 
rewrite would have to happen anyway ....

Of course, being able to understand Carl's R3 architecture might 
be REALLY inspirational, so I would not discount that option :-)