Hi Paavo,

Just tonight, a reader from my site wrote to say that in the past, all =20
the documentation he had previously read had led him to misunderstood =20
the syntax of a simple 'if expression, originally thinking that the =20
conditional evaluation needed to be placed in brackets (which made =20
every condition always evaluate to true...).  Even things like that =20
can be tripping blocks for newcomers, so I agree that obfuscated code =20
can be totally unfathomable.

With that said, it is important to recognize common code patterns and =20
Rebolisms that are fast and short in expression, even if they're hard =20
to read at first.  I don't know that common (more difficult to read) =20
syntax necessarily needs to be avoided and replaced with obvious =20
patterns.  (If they're common, they _should_ be internalized).  =20
Instead, what I missed in the beginning with Rebol was more documented =20
examples ... I mean the kind of detailed, line by line documentation =20
needed to understand each expression, as well as the overall structure =20
of a script.

I love the idea of maintaining a collection of useful, classic =20
programs that demonstrate common techniques and solutions to typical =20
problems and situations.  Having a clear English language =20
translation/explanation of any code, no matter how unreadable, can =20
make it much more friendly to digest and internalize.  If I were to =20
put forth the effort, instead of rewriting examples to make them more =20
readable, my suggestion would be to just write lots of more detailed =20
explanation - LOTS more.

Anyone interested in making Rebol more accessible to others could =20
potentialy help by adding detailed documentation to scripts that are =20
in the public domain.  Anytime someone has an "aha" moment, =20
understanding how a certain expression or code pattern works, in the =20
example where it occurred, notating that realization would undoubtedly =20
help others down the road...  I'm sure that line-by-line explanations =20
of more code from rebol.org would help average new users dramatically. =20
  Anyone can help with that sort of thing - and the wiki sites are =20
great resources for creating repositories of analyzed and notated code.

If anyone is interested in starting a project like that, we just need =20
to copy over a few scripts to a chosen wiki, and add our own comments. =20
  As an example, I put Frank Sivertson's Rebtris, up at =20
http://en.wikibooks.org/wiki/REBOL_Programming/Examples .  I wonder =20
how long it would take to get such a script completely notated...


Quoting Paavo Nevalainen <[EMAIL PROTECTED]>:

>
> We have a tendency to produce code, which is ultimately fast or short
> in expression. It is so since we maybe compete a bit which each other,
> which may lead to bizarre code from the beginner's point of view - and
> from the point of view of any non-reboled person. And even a "personal
> programmer" have to cope with others nowadays....
>
> So, it could be useful to rewrite some "classics" again readability in min=
d.
> Some examples could come in two versions, prototype or documentatory
> version, and optimised production code.  I could help in that process, whe=
n
> R3 is ready (I know I am a simpleton, so obviously I am not the one who
> produces the optimised code ;) ) .
>
> --
> To unsubscribe from the list, just send an email to
> lists at rebol.com with unsubscribe as the subject.
>
>


-- 
To unsubscribe from the list, just send an email to 
lists at rebol.com with unsubscribe as the subject.

Reply via email to