[REBOL] Re: InformIT.com Articles Introduction to Water™ A New Native We
From: gerardcote:sympatico:ca at: 30-Aug-2002 14:12
Hi Robert,
You wrote :
> Hi, might be the message was to cryptic... I didn't got in the first
> reading.
-------------- Gerard -----------------
You are completely right Robert, since I have some difficulty to cope with my English,
I often have to take long and winding roads
to express my ideas, really longer than it could take in FRENCH, for example.
Sorry about my muddy verbose English - I will try to be sharper and more concise in the
future :-))
------------- /Gerard ----------------
> Well, I'm working on a rebol-framework just now. It includes
> the ideas of associative data model, nakedobjects ideas etc. I have an
> idea about the GUI already, I have the basic building blocks running.
>
> I have a first draft.
-------------- Gerard -----------------
Nice , and I also suspect that Andrew is doing something like this too, based on similar
foundations.
-------------- /Gerard ----------------
> But I'm not sure if releasing it makes a lot of
> sense at the moment. I'm an absolute fan of community development, but
> there are two sides of it:
-------------- Gerard -----------------
I agree with you pov relative to the state the project has to worked before it will be
of some usefulness to the community. I admit
we will have to wait and see until someone whatever it will be you or anybody else or
a team - will have a Useable toolkit to share
with the rest of us and for getting some advice to help enhance the product together.
As I already offered previously, I'll be glad
to participate when you'll decide it is ready to be done. Until then I continue to study
REBOL and other IDEs that should be of help
in a near future - even if it is solely for the purpose of brainstorming before DESIGNING
and HARD CODING a first too bad solution .
-------------- /Gerard -----------------
> 1. I need to make progress as I don't want to work on a never ending
> project. I concentrate on doable ideas etc. Vision is good, but reality
> needs a solution.
-------------- Gerard -----------------
I agree with you too on this point. I often send up-front many ideas without having a
good estimate of all the work that has to be
done to get all of them functional. I appreciate to have some "realistic" guys like you
and most of the ML members that are able to
SAY : NO MORE when it has to be done ... even if this implies that I must wait for another
more or less time period. At least I'm
not staying inactive on my side either !!!
-------------- /Gerard -----------------
> 2. Getting the same understanding and agreement about how to continue
> etc. is mostly impossible in a community process. This leads to the
> danger of forks for the different concepts. IMO this is the worst thing
> that can happen, we need to work on one framework.
> Therefore I think it might be better to develop the idea up to a useable
> level and release it than. I'm hoping that enough of the community will
> jump on and follow this direction than... I don't know if it will
> happen, we will see.
-------------- Gerard -----------------
But here my pov is simply a matter of not going too far in a CUL-DE-SAC (dead end) while
it could be done otherway before if some
other external party had just eyed over your shoulder when it was time... and the time
is before the HARD CODING begins. But I
understand that with REBOL we can try and test better and more affordably than with other
Programming Languages - generally speaking
of.
Even Carl and RT understood the fact you are submitting and to some point I agree with
you. The only reason I am pushing against
this so much is the time me have to wait to get a lot more - when compared to other communities
like the Python one and more
recently the Ruby one.
Just to give an example of what is going on the side of the latter - Many of ML members
want and have to wait for a new REBOL
ActiveX Scripting Engine for MSIE. That is an other one than the Jscript and VBScript
ones offered by MS. I already seen that Ruby
has its own, mostly due to the fact that Ruby is not as closed as REBOL is for now.
May be one solution would be RT just let a Hook to attach REBOL to other Languages, without
having to get the PRO or COMMAND
version.
I am currently looking at some toolkit ( or I could even write one myself with some more
time involved ...) that could help me to
develop a WRAPPER around the REBOL engine for using it as another ActiveX scripting Engine.
Borland's Delphi looks promising for this kind of hack even if I am not really a system
programmer. But this also is asking a lot of
time while letting some REBOL HOOK open for external use would be a lot shorter for Carl
and his team. But he has his own reasons
for not doing so and I can't necessary be waiting for him.
Another field of interest is having an IDE develpped in REBOL for REBOL to help accelerate
and standardize some develop-test-run
cycle for REBOL scripting. Instead of always reinventing the wheel, we could be using
some REAL templating system to start with when
developing CONTEXTUAL apps. Similar in concept to the DIALECT one but could be used in
a more VISUAL form than now. It's becoming
more and more time consuming to look and study new approchaes and new code every time
for newcomers when some ppl already stated on
some very useful way to get things done. This knowledge should be better used than it
is now - even if I pertinently know that
everyone is giving the max for the others in this ML group. I reverently follow almost
every thread - I am able to understand - on
this ML and I really appreciate the spirit and time that is shared by every member. And
I know we already have some library and a
ZINE too - since I study code from both of them but time being what it is -
In fact I would like to be part of some collective movement that would have as a goal
to shorten the REBOL learning curve and
enhance the App. Dev. process if it can be done - and I am sure it can.
In a similar way, I am also currently working - shen free time is left - to find a way
to fork a PHP or PERL script that could launc
h a REBOL engine for doing CGI on ISPs HOSTs that don't want to support REBOL CGI with
their APACHE Web server. My own Linux based
ISP doesn't want to configure its APACHE server for letting me do my CGI with REBOL and
I want to show him it's easy and safe to do
it - even the way I plan to do it.
I also already met and endorsed the proposed eREBOL solution ( which would act as a replacement
for PHP) and I admit it can be used
for some useful tasks related to WEB / HTML / XML generation and maintenance. But then
I come back by suggesting with a simplified
DIALECT version of the Water language I submitted to the ML, a couple of days ago.
May be the PHP /PERL approach will not be very efficient but for some small load it could
be sufficient for me and some eventual
customers.
Another alternative is to get its own Web Server and act like another ISP but not everybody
wants to be in charge of setting-up and
managing such a system for occasional or even day-to-day operations.
If someone want to help on any of these threads, be kind to do it. Everybody's assistance
is accepted ;-)
-------------- /Gerard -----------------
> > Who else want to be
> > a part time coordinator/and/or collaborator for this task and
> > help start this App Framework project - nobody has to start or finish it
> > alone but someone must be in charge for the first draft on
> > which we'll work after ??
>
> I can do this. But I can't spend a lot of time etc.
-------------- Gerard -----------------
I thank you for this implication Robert. Hope it will be as successful as you hope it
will. Have a share with us as soon as you'll
think it's OK for discussion or a try.
-------------- /Gerard -----------------
>We have a trainee that will start to program some script for me in Rebol. But he will
need
> some time to ramp-up and doing framework development won't be his job in
> the next month. He can test etc. But not more.
>
-------------- Gerard -----------------
Looks like myself at this moment ... If you want to get more testing help, I have somewhat
more free time this fall.
-------------- /Gerard -----------------
Regards,
Gerard
P.S. I think you're on the good track when usin the ASSOCIATIVE MODEL and NAKED OBJECTS
as the basis of your FRAMEWORK
