[REBOL] Re: starting to be really late!?
From: petr:krenzelok:trz:cz at: 11-Jan-2004 18:55
Karl Robillard wrote:
>Hey Ed, that's an excellent summary of how Rebol differs from other mainstream
>languages.
>
>Everything else in this post I have said before, but I like to restate my
>thoughts whenever this topic pops up, as Robert notes that it does, every few
>months or so.
>
>I believe the artificial limits placed on Rebol by RT is the problem and agree
>that the strategy is wrong. I don't want another platform and Rebol is
>doomed to failure if it insists on trying to compete with all the existing
>ones. I want Rebol as a tool to bind my existing platforms and applications
>together.
>
And I have to ask why do you want to limit rebol in becoming kind of
Arexx universal glue, when it can actually provide much more usability?
I can see the main rebol concept as very good - the one of rebol block
and mixture of code = data (even binary ones) principle. My opinion is,
that there is not so much effort needed to push rebol in certain areas.
How do you want to judge what aproach is right? You want kind of arexx,
someone else might be interested in Authoring tool (right cyphre? :-).
But then I often hear that that why to compete with Flash etc. kind of
argument and my opinion is - that is not true. E.g. I have no intention
(limited by lack of free time) to learn FlashMX to become fluent in such
environment and besides that I think that some nice changes to View
would eliminate such need.
The problem of past of RT was imo their investors. I do remember when RT
switched to IOS (Express back at that time) kind of strategy. I really
disliked it. They had no corporate experience imo and what is more, even
IOS development stopped while there is so much what can be done to make
it better. While the glory days of RT employing several good coders is
over, I think there can be way how to handle current situation:
- community cooperating with RT - View 1.3 project is good start. I
think that all ppl there are properly focused, no extra noise on
particular channels etc. The difference between 1.3 effort and last year
effort is clear - Carl's almost daily involvement, while in the past
Carl appeared just once per 2 or 3 months to check. Other difference is
often beta releases for betatesting ....
- the way to go imo - RT concentrates on addition of general engines.
E.g. - why to wait for RT to add particular functionality, feature, if
we could add it ourselves? And yes, I mean addition of virtual machine,
or at least specific area dialect engines for fast manipulation (e.g.
working with image data). That way we could add additional effects
ourselves, or in the case of RVM, code some additional ports/drivers to
other environment ourselves, while stying fast - the advantage agains
linking to libraries is obvious - rebol app would stay self-contained,
RT would be freed from I-want-this-he-wants-that kind of feature
requests. In fact, talking to Carl few days ago, he mentioned he thinks
of addition of such principles into rebol more and more. That is imo a
good sign.
- also - Carl said that he could add more functionality into rebol, if
we would help to find C source codes, .e.g. that he would be able to add
RGB24 convolver in less than day probably. GIF saver and wav loader were
also mentioned in that respect. So, it is also upon us. Similar thing I
remember about .zip, .arj., .rar code, where the idea was to treat them
like ordinary directories or to create schemes for them, something like
zip:// ...The offer was made, but we (the community) failed to deliver
... so ...
There are of course other things ... newer GC, async networking sitting
in separate dev. tree, not yet merged, things like more granular event
system, language extensions, direct draw modes etc. But - we should keep
in mind we are short on resources. But I do believe, that if we work
with so much dedication as can be seen in the case of View 1.3, the
things will get better and better. Of course it would be cool to be able
to clone ppl like Romano, Volker, Gregg and many others, who do
wonderfull work for 1.3, but the situation is as is :-)
-pekr-
