Development rebol scripts on Linux
[1/13] from: fergus4:bellatlantic at: 15-Nov-2007 13:33
I like Linux but I do not enjoy Rebol programming on it. Mostly due to the
lack of rebol console on Linux.
When I run a program from an icon and it fails you do not get the error info
as no console pops up. I'm not a big console guy so I tend not to start
programs from the linux command line.
Any suggestions?
Alan Macleod
[2/13] from: ale870:gmai:l at: 15-Nov-2007 22:05
Good point.
It's true. Rebol console linux is not so user-friendly like the one in
windows.
I don't know if R3 has a better Desktop. Can somebody confirm something
about R3?
On Nov 15, 2007 7:33 PM, Alan <fergus4-bellatlantic.net> wrote:
> I like Linux but I do not enjoy Rebol programming on it. Mostly due to the
> lack of rebol console on Linux.
<<quoted lines omitted: 7>>
> To unsubscribe from the list, just send an email to
> lists at rebol.com with unsubscribe as the subject.
--
//Alessandro
http://sguish.wordpress.com
http://laccio.wordpress.com
[3/13] from: tim-johnsons:web at: 15-Nov-2007 18:14
On Thursday 15 November 2007, Alessandro Manotti wrote:
> Good point.
> It's true. Rebol console linux is not so user-friendly like the one in
> windows.
> I don't know if R3 has a better Desktop. Can somebody confirm something
> about R3?
Yup. Rebol console on linux sucks. So sad because of the readline resources
available. I use emacs, which provides 'comint' or 'inferior process' mode
that enables the following:
1)Running rebol *inside* of the emacs editor
2)Sending code from a file directly to the rebol interpreter for evaluating
3)Loading files
4)*AND* writing code on the command-line (when running inside of the
interpreter) make use of all of the emacs editing features.
The mode is available. I wrote it from scratch and it is tailored to my own
needs, but should work for other users and should be easily modifiable.
tim
-- Binary/unsupported file stripped by Ecartis --
-- Type: image/png
-- File: rebol-emacs.png
[4/13] from: tim-johnsons:web at: 15-Nov-2007 20:19
On Thursday 15 November 2007, Tim Johnson wrote:
> -- Binary/unsupported file stripped by Ecartis --
> -- Type: image/png
> -- File: rebol-emacs.png
That image is just window dressing, if you want to
see it let me know and I will send to you directly.
tj
[5/13] from: santilli:gabriele:gm:ail at: 16-Nov-2007 12:42
2007/11/15, Alan <fergus4-bellatlantic.net>:
> When I run a program from an icon and it fails you do not get the error info
> as no console pops up. I'm not a big console guy so I tend not to start
> programs from the linux command line.
Depending on which file manager / desktop you are using, you should be
able to right click on the icon and select "start from terminal" or
something like that.
HTH,
Gabriele.
[6/13] from: tim-johnsons:web at: 16-Nov-2007 7:50
> 2007/11/15, Alan <fergus4-bellatlantic.net>:
>> . I'm not a big console guy so I tend not to
> > start programs from the linux command line.
Another way might be to do your testing from a script,
set up your file associations so that the script is run
on a mouse click.
Example:
you have a file with code inside of functions.
You execute the code that you want to test from a
function - let's call it 'test.
You put "test" as the last line in the file. Then
you run it from a file manager - OR you could
also run it as a CGI script. I do that a lot when
I want to capture a lot of output.
You can then test your code from whatever editor
you feel most compatible with. Emacs is a bear to learn.
Nothing beats the interactivity of a console tho'.
I am (I think) the first person to develop an interactive
class on programmer using rebol. My great frustration
was that the students students wanted the "bells and whistles" -
i.e. drag-and-drop, all kinds of windows popping up and
plenty of other stimulus and were not able to concentrate
on small discreet portions of code.
<rant>IMHO this is one of the reasons that my
county (USA) is losing programming jobs - to much
desire for stimulus on the part of younger generations.
</rant>
Gotta love the code....
Anyway, I hope we are being of some help to you Fergus.
regards
Tim
[7/13] from: carl:cybercraft at: 17-Nov-2007 20:01
On Friday, 16-Novenber-2007 at 7:50:39 Tim Johnson wrote,
>Nothing beats the interactivity of a console tho'.
>I am (I think) the first person to develop an interactive
<<quoted lines omitted: 7>>
>desire for stimulus on the part of younger generations.
></rant>
The kids are aways right Tim - never forget that! ;-)
And they have a point. They might not know how to say it, but they'd instinctively know
that it'd be much easier and faster to create a GUI using point & click than by typing.
A mouse & GUI will have been the only way 90% of them have ever got things done on a
computer, since that's been the default interface on personal computers for over 20 years
now.
And point & click should be how you drag the kids into getting interested in coding.
Use a top-down instead of bottom-up approach. Start with designing the interface (using
drag-and-drop) and move from there down into the coding and making the interface work,
not the other way round. (That there may not have been a suitable REBOL tool around
for this approach isn't the kids' fault - they were still right.)
Oh, and you're losing programming jobs because those from Asia are much, much cheaper.
The observant kids might've noticed this too and realised that coding in the future
might be no more lucrative a job than being a factory or farm worker is now.
(When J. K. Galbraith was asked where he got his work ethic from, he used to always reply
with "You've never worked on a farm." Was reminded of this quote when a certain someone
said he wouldn't want to be a professional vineyard farmer... The irony of this is that
down here in NZ, which is basically one big farm, vinyards are seen as farming-lite;)
-- Carl Read.
[8/13] from: tim-johnsons:web at: 17-Nov-2007 10:59
On Saturday 17 November 2007, Carl Read wrote:
> On Friday, 16-Novenber-2007 at 7:50:39 Tim Johnson wrote,
> >Nothing beats the interactivity of a console tho'.
<<quoted lines omitted: 15>>
> way 90% of them have ever got things done on a computer, since that's been
> the default interface on personal computers for over 20 years now.
It is not a matter of being *right* or *wrong* it's about a mindset.
One has *got* to be able to focus on algorithms.
BTW1: I came to rebol from C++ builder with a background in combining
drag-and-drop with coding.
I'm well aware of the economics. JKG might have suggested that the world
needs a global monetary standard to level the playing field, but that
is beyond me.
I stand by my contention that you have to be able to free onself from the
temptation of high stimulus to really use your imagination.
Look at what Einstein did - and a lot of what he came up with happened
when he was hiking in the Alps.
I grew up on a farm,, most of my family still farms and I've done my share
of bushwhacking and subsistance living. Amazing ideas have come out
of the austere life led on the central plains of the North American
continent. And other rural open environments as well.
The local electric utility routinely train their CAD staff in a special class
that emphasis the keyboard. They claim that it gives a 30% increase of
productivity over relying solely on the pointing device.
I use emacs for coding rebol. I primarily use the keyboard, with the mouse
as needed and where it works best. Comparing this environment to coding
in C++ builder is like comparing a sportscar to a truck. Of course trucks
have their place but since I am a single coder instead of part of a team, this
works best for me.
Nice arguing with you. :-)
Did you know that J.K. Galbraith and William F. Buckley were best friends?
WFB describes JKG as his "dearest friend" - think of the arguments they
must have had.
Regards
Tim
[9/13] from: btiffin:rogers at: 17-Nov-2007 15:59
Gentlemen; I don't understand. :)
This has not been my experience at all. Maybe it's because I am more of a fan
of the CLI over the GUI when it comes to development. Actually I should say
a nice CLI inside a nice GUI.
KDE, Konsole and Kate. Beautiful development environment. (It's not
official, but scraping the net I found a Kate REBOL syntax highlighter).
http://lists.kde.org/?l=kwrite-devel&m=111834309909165&w=3 by Roland Hadinger
(completely manual operation getting it installed ... pulled out the XML and
then did the whole reading up on Kate extensions thing to get in plugged in
properly...but it's wonderful).
Konsole for tabs, (minimum 4 REBOL consoles running 24/7. One tab for typing
HELP and SOURCE and thes/def/spell (from dict-demo.r in rebol.org) One for
the Viewtop and wordbrowser, the Desktop Librarian and VID editor, one for
testing View/VID/RebGUI snippets and one (and then more) for running the days
development work-in-progress.
Kate for the pretty colours and project management through sessions, vi for
getting quick things done and tagging (thanks to rtags) for the whole
cross-reference thing. Kate's built in terminal keeps the working directory
in synch with the file being edited. S'all good. Kate even detects when vi
(any external edit) has changed the file and offers reload/ignore.
REBOL staying in the startup terminal is nice. Getting Cygwin running on
Vista (Cygwin console mode...not Cygwin/X) is the only thing keeping me sane
in the Vista environment. rxvt and vi rules. :) (emacs is ok and all, but
I think Richard Stallman has eight fingers on each hand). Bill Joy is
probably a thumb typer like myself.
To all:
So if you haven't yet; dump Windows...snag a nice GNU/Linux distro (Debian),
run KDE/Konsole/Kate/REBOL and you may never look back. If you are really
stuck with Windows, then Cygwin your way to pretending to have a real OS
running. :) Nothing like the power of the Unix CLI ... the best IDE ever
invented.
Note: I'm old; (well past my Carousel date in Logan's world) ... you're
mileage may vary.
Cheers,
Brian
On Thursday 15 November 2007 16:05, Alessandro Manotti wrote:
[10/13] from: ale870:gm:ail at: 17-Nov-2007 22:19
Hello guys, this discussion is very interesting and I think all of you are
right, and all of you are wrong.
Why?
Because there are two important points to remember:
1) One makes a program to gain money or to enjoy itself? If you want to gain
money, you need to maximize the results and minimize the time to do them:
get the best results with the minimum effort, stress. If you make a program
to enjoy, then do everything you like, since everything satisfy your mind
and your soul is good in this case;
2) Does not exist a programming language to make everything: every
programming language best fits specific targets. Please tell this to
everybody that say that drag'n'drop is better (but better than... ?) Well,
try to create a kernel module (or an entire O.S. using a drag'n'drop
system... no. In that case you need an efficient IDE, text based, and... C
language (or C++). No GUI, no buttons. You need only an IDE, a lot of
documentation, and a good brain!). If you want to create a rich client
application, then you could need a drag'n'drop, a RAD tool, documentation,
and a good brain!
I think it is not important which language you learn, or which tool you use.
A good programmer (analyst/programmer) needs a good method, a strong
preparation, a good brain to analyze and solve problems, a good approach to
the different kind of programming (procedural, object oriented, functional),
and... a bit of lucky (just to be lucky to use a compiler without too many
bugs! :-) )
On Nov 17, 2007 9:59 PM, Brian Tiffin <btiffin-rogers.com> wrote:
> Gentlemen; I don't understand. :)
> This has not been my experience at all. Maybe it's because I am more of a
<<quoted lines omitted: 72>>
> To unsubscribe from the list, just send an email to
> lists at rebol.com with unsubscribe as the subject.
--
//Alessandro
http://sguish.wordpress.com
http://laccio.wordpress.com
[11/13] from: tim-johnsons:web at: 17-Nov-2007 15:45
On Saturday 17 November 2007, Alessandro Manotti wrote:
> 1) One makes a program to gain money or to enjoy itself? If you want to
> gain money, you need to maximize the results and minimize the time to do
> them: get the best results with the minimum effort, stress. If you make a
> program to enjoy, then do everything you like, since everything satisfy
> your mind and your soul is good in this case;
I work for the love of it and for the money. Both are equally important.
> 2) Does not exist a programming language to make everything: every
> programming language best fits specific targets.
Right on. And paradigms be damned. Pragmatism trumps 'em all.
> I think it is not important which language you learn, or which tool you
> use. A good programmer (analyst/programmer) needs a good method, a strong
> preparation, a good brain to analyze and solve problems, a good approach to
> the different kind of programming (procedural, object oriented,
> functional), and...
I heard an interesting talk by one Carl S (yes that winemaker guy :-))
He described the human brain as a very slow but massively parallel
computer. That makes a link to the experience that I have had, where
solutions present themselves apparently spontaneously in my brain,
but probably as a result of the "slow and massively parallel" phenomena.
One of my pragmatisms is to put myself in situations *away* from the
computer where I can let the parallelism work...
> On Nov 17, 2007 9:59 PM, Brian Tiffin <btiffin-rogers.com> wrote:
> > Gentlemen; I don't understand. :)
> > :) (emacs is ok and all,
> > but
> > I think Richard Stallman has eight fingers on each hand).
I don't use emacs because of Richard Stallman, but because of
the elisp environment. In a sense, elisp has many of the same
roots as rebol. RMS was in Alaska, gave some talks at the University,
rubbed so many people the wrong way that there was talk about
taking him out in the woods and leaving him for the bears.
> > Note: I'm old; (well past my Carousel date in Logan's world) ... you're
> > mileage may vary.
Well, I'm 58 and have been coding for 20 years. I went back to college at
the age of 38 to study computer science after working all my previous
life with my hands. Most of my fellow students were 15-20 years younger
than me and I left a lot of them in the dust. Not because I was a fast
learner, but because I had learned to handle failure and when I failed,
I just picked myself up and tried again.
Maybe I like emacs because I still *do* work with my hands. :-)
BTW: I use vim for all of my adhoc editing and system work.
As the default editor for MC.
MTCW
tim
[12/13] from: ale870:g:mail at: 18-Nov-2007 13:44
Hello,
I heard an interesting talk by one Carl S (yes that winemaker guy :-))
> He described the human brain as a very slow but massively parallel
> computer. That makes a link to the experience that I have had, where
> solutions present themselves apparently spontaneously in my brain,
> but probably as a result of the "slow and massively parallel" phenomena.
>
Yes, this is what the study of the brain are reaching: it seems the brain
may do no more than 5-10 operations per second (compared to a computer-like
approach), but the brain has A LOT OF memory of a computer, and its database
is really more efficient that any existing database. Furthermore, database
in the brain is optimized to store analogic information, and "multimedia"
data, like images (well, video :-) ) sounds , smells, etc.... WIth our
actual technology, we cannot do anything similar. More: I don't understand
why human people want to simulate the mind functioning... I use a computer
when I want to make a calcuation that should be perfect... imagine a
computer that makes calculations like a human mind: every calc should be
right (but you are not sure that it is so!). Who could use a such
calculator?!?!
A.I. is great, but we don't confuse A.I. with human mind functioning.
Well, I'm 58 and have been coding for 20 years. I went back to college at
> the age of 38 to study computer science after working all my previous
> life with my hands. Most of my fellow students were 15-20 years younger
> than me and I left a lot of them in the dust. Not because I was a fast
> learner, but because I had learned to handle failure and when I failed,
> I just picked myself up and tried again.
>
You have all my respect! I'm 37 and I don't think I could be so "strong" to
make a such huge effort. My congratulations for your great job! And for your
great results!
On Nov 18, 2007 1:45 AM, Tim Johnson <tim-johnsons-web.com> wrote:
> On Saturday 17 November 2007, Alessandro Manotti wrote:
> > 1) One makes a program to gain money or to enjoy itself? If you want to
<<quoted lines omitted: 48>>
> To unsubscribe from the list, just send an email to
> lists at rebol.com with unsubscribe as the subject.
--
//Alessandro
http://sguish.wordpress.com
http://laccio.wordpress.com
[13/13] from: carl:cybercraft at: 19-Nov-2007 12:10
On Saturday, 17-Novenber-2007 at 10:59:28 Tim Johnson wrote,
>> ><rant>IMHO this is one of the reasons that my
>> >county (USA) is losing programming jobs - to much
<<quoted lines omitted: 10>>
> It is not a matter of being *right* or *wrong* it's about a mindset.
> One has *got* to be able to focus on algorithms.
But before that they have to be persuaded that they'll like programming. It's like the
old problem of throwing Shakespeare at students before they're ready for him - they just
end up hating Shakespeare. You can't avoid having to learn how to develope algorithms
if you want to create programs. But if kids start with designing GUIs first, by the
time they're forced to drill down to the coding level to make their GUIs work, they'll
actually have a real incentive to learn how to program - namely the getting their GUIs
to work.
> BTW1: I came to rebol from C++ builder with a background in combining
> drag-and-drop with coding.
<<quoted lines omitted: 5>>
> Look at what Einstein did - and a lot of what he came up with happened
> when he was hiking in the Alps.
Well of course - nothing acting on his senses up there...
But by your argument, he should've stayed home staring at nothing but flat surfaces with
characters on.
That said, there's also fundamentally different kinds of people about, and I'm apparently
a very visual type of one. There's tests online which measure such things, and I always
come out as a heavily visual kind of thinker, as apposed to those who think with words.
(Which I don't understand at all!) Your milage may vary...
> I grew up on a farm,, most of my family still farms and I've done my share
>of bushwhacking and subsistance living. Amazing ideas have come out
>of the austere life led on the central plains of the North American
>continent. And other rural open environments as well.
Meaning you're not far removed from those Asian coders who're very well aware what subsistance
living means. Most of the young in the West are a few generations removed from knowing
what a day or a week behind a plow is like though, so fear of returning to that won't
motivate them. They need incentives.
>The local electric utility routinely train their CAD staff in a special class
>that emphasis the keyboard. They claim that it gives a 30% increase of
<<quoted lines omitted: 4>>
>have their place but since I am a single coder instead of part of a team,
>this works best for me.
But you're not comparing what you should be. You should be comparing REBOL with a hypothetical
REBOL Builder that's as big a productivity gain as REBOL is over C++.
I've never used a GUI designer I liked, but that doesn't stop me knowing a well-written
one would be boon to any language that supports a GUI.
>Nice arguing with you. :-)
>Did you know that J.K. Galbraith and William F. Buckley were best friends?
>WFB describes JKG as his "dearest friend" - think of the arguments they
>must have had.
I don't know Buckley. And a check in Galbraith's memoirs finds only one mention of him
in passing. The book's over 25 years old though, so plenty of time for a friendship
to form.
-- Carl Read.
Notes
- Quoted lines have been omitted from some messages.
View the message alone to see the lines that have been omitted