What's Wrong With REBOL?

[Edoc]

I think a dialect or Ashley's solution is the pragmatic approach.

My guess is that a refinements change to the /Core interpreter could have very deep consequences and potentially break the majority of available scripts.

Pure conjecture: I don't think REBOL/Core is designed with an ounce of "slack" in it. For the sake of compactness, /Core is coded as tight as a monkey's fist knot. Due to dependencies, changes can have cascading effects throughout the rest of the system. Before a change is made, the impacts must be carefully ruminated, cogitated and meditated upon before the first keystroke is made.  

[Gregg]

A dialect, or more than one, is definitely the way to go IMO. I think we have 99% of the tools we need right now to try out improvements to things like PARSE. If we find something really good, that everyone--including Carl--agree should be in REBOL, then RT can add them.

We're still only seeing the tip of the iceberg WRT what REBOL is all about (again, IMO) so I'm *very* leary of suggesting changes to the core syntax. Carl spent a looooooooooooong time designing that and is one of the few people--maybe the only one--who has looked 5 or 20 years down the road with the vision for what it all means in the context of REBOL's true calling.

[Paul]

Well said I agree Gregg.

Paul

[CarlRead]

One serious problem I have with the REBOL syntax is the way refinements work.

The refinement appears next to the command, but the values associated with that refinment appear after the other paramaters.

I think something like the following would be so much easier to read:

CODE

db-select * my-table /where [id = 1]
 

REBOL, once learnt, is an easy language to write, but I agree it's not too easy to read REBOL code written by others, especially for beginners.

Still, I don't think seperating the refinement name from it's word would improve readibility.  For instance, how would you sort this out...

Code:

func1 func2 some-code /ref1 more-code

if both functions had a refinment called ref1?  It would force you to use parens, resulting in something like this...

Code:

func1 func2 some-code /ref1 more-code; ie, func2/ref1

func1 (func2 some-code) /ref1 more-code; ie, func1/ref1

Also, I think the interpreter would require parens even if both functions didn't have the same refinment name.

[jj]

why aren't "hackers" using Rebol I wonder? should only open source wares have grass-roots support?

it's so portable both among platform and in most literal sense, I mean I have an USB memory key loaded with Rebol and some useful scripts. coolest thing!
can we do this with python or perl?

how much do those college professor rate Rebol? how much chance do we have for Rebol being introduced in Computer Science Depts? Industry and / or Basic aside, college is the place for most people to get idea about computing...etc

[leke]

In responce to the original question...

Taken from REBOL Essentials:
‘In 2002 REBOL was even listed as nominee for the Webby awards for technical achievement, nevertheless it's still rarely known.’
Viktor Pavlu

Nobody I know (interested in programming/the internet) has ever heard of REBOL. In my opinion this is why REBOL has not ‘set the world on fire’.
Does a new marketing strategy need to be implemented?
At the very least I think it needs to do the rounds in Computer Magazines CDs (PC PLUS for example).


Larry Wall put this language in the category of a ‘Cute Toy’. If this is true, there are millions of programming hobbyists out there who would love to play with it! I found REBOL while literally going through Googles ‘Programming Languages Directory’ (a kind if programming toy box so to speak).*
REBOL is my Buzz LightYear Doll and I wouldn’t swap it for any other toy, except maybe a girlfriend;)

*Saying this though, I never got past ‘R’ I wonder what lay be yonder?