AltME groups: search

If the rpm is for your redhat and has a 2.3, install it? after lots 
of backups maybe, i guess if the lib fails all fails. or unzip it 
and put the library somewhere else and use LD_LIBRARY_PATH. Do you 
have a link to the package?

I mean, it's a system running in a data-center, and I really have 
to have this system up and running.

Robert: it might be easy to pick a new box and move data. But, the 
effort to make REBOL  work is probably much smaller.

Volker do you need RH 7? Else upgrade to Fedora 4... plug and play 
actualy, Not my Linux Linux flavor, but Fedora hits close to RH 8.. 
Or else look for fedora 3 ?? RH 7 i guess.. You install it in less 
then 15 minutes !!! just did it today..

I suppose perhaps the alternative is to build a boot floppy and install 
over FTP from the box that has the dvdrom. It's been years since 
I had to do that. Too bad Debian doesn't provide a little script 
to cut CDROMS from the DVD. I've got to guess I'm not the only person 
with this issue.

And you have to wonder why they need a 10GB distro. "Just give me 
the main system and libs please" - no bloatedness, thanks.

Another, easy, option would be MEPIS Linux. It's much closer to Debian 
than Ubuntu, although based on a newer version than the stable version. 
And very easy to install

Well as I said befor... Slackware and debian are the only two with 
unmodified lib and kernel structures.. and are great development 
envs...a pitty though about that dvd from debian..

people who startout using Ubuntu, Fedora or Redhat find themselfs 
In an Linux world that not even default with i.e. Disk-Structures 
on Tru64 or Solaris or HPUX.. while using other Linux distributions 
from around 1995 they stick with the UNIX system (IV|V) layout..But 
when it comes to development I can emagine to make a choice between 
"package handling" and "environment handling"... Then I would stick 
with debian..just has a better package manager...where slackware 
is a little more the "do it yourself package manager.." still 'tar 
does wonderful things.. for the user...

Knoppix can be installed and is is then a debian, so the rest can 
be apt-getted. damn small linux claims the same. Kanotix is a knoppix 
specially made for that. no guaratees..

However, all the newer ones give me problems. Here is why: I used 
to be able to build Libc6 and Libc5, and that was all that was necessary. 
 However, now if I build on Ubuntu and run it on Mandrake, it crashes... 
and every other combination.

So, I studied the situation a bit, and found that Debian is really 
the meat behind a lot of these distros, and someone I trust suggested 
that Debian made a better dev platform and would provide the perfect 
build standard for us.

I searched far and wide on the net, and found no info at all on how 
to do this.  I even wrote a REBOL script as a mini Linux package 
manager, because boiling down 10GB to .7GB is no small task.

I also built the floppy set, and figured I would use my own local 
Debian archivel site from a different local machine.  So, mounted 
the DVD on one system and published it with HTTP, and the floppy 
boot worked, found the host, but the installer rejected the archive.

So, it was crash and burn.  I have better things to do right now 
that screw around with Linux. So, it's on hold.

exactly - I know Carl talks about principles ... but why to loose 
your time with fighting to get things stripped down to CD? I would 
go and bug DVD drive ....

To recap, most of the newest "user friendly" distros are based on 
Debian, so Debian is the common denominator for them. If you add 
to the list SuSE (commonly used in Europe), RedHat (and therfore 
Fedora), and Slackware, I would suggest to stick with these for distros 
and build specific version of Rebol for them...

the problem is that normally a person makes a program and then multiple 
maintainers make sure it runs under certain distros

Yes, and look how nice open source has become these days.  It is 
a huge mess.

Carl, a suggestion: why not make an agreement with trusted Rebol/Linux 
users and delegate the build on specific platform. Kind of very restricted 
and contractuel "Open Source"?

Anyway, I will delegate this task now to someone here, and start 
taking the box apart. My last choice. Always.

I fully agree Carl.. Long life the "spaggeti Open Source Community" 
....Windows rules ;-) on some parts...it realy does!.... (and Im 
a fanatic Unix/Linux user..)..I was struggling for 2 days with my 
DVD drives for a simple 1-on-1 copy (none data) ..and none of the 
Linux packages where able to correclty copy DVD's of  my format.. 
Jumping back to windows XP (i though ..iekss) ..seeking some obscure 
sites and found a hell of a tool that copy's everything.."dvd decrypter" 
(EOL btw but free..) And that something linux community did not build 
yet...Awfull.. Well im off taking care of some windows backups ;-) 
It still stays windows ;-)

I do not want to download anything from the net, nor should that 
be necessary..
 I think thats the main problem, specially with debian..

 and found that Debian is really the meat behind a lot of these distros

 thats right. Which is why i suggested knoppix. Which is a debian 
 modified to run from cd without needing harddisk. Which was lthen 
 modified to dump itself to hd as debian-installation. Which is a 
 nice way to install debian IMHO.

so I tried 2002 version of my friend's Knoppix and I was impressed 
(but I miss my Total Commander, huh :-)

I am willing to give new Knoppix or Ubuntu a try - what do you suggest 
please for small, 1CD intro, which is able to easily mount your NTFS 
partition and let's you to delete windows, program files, or simply 
to back-up some files and does not screw your filesystem?

Well, I have tried captive ntfs and it worked well (no accessive 
testing, though). It uses your NT/XP drivers ... http://www.jankratochvil.net/project/captive/

I think part of the "mess" of OSS is the perception that it is a 
single movement, a unified army of developers. That is completely 
wrong, even though some of the louder proponents of open source work 
hard to make you believe exactly that. Even folks who understand 
the truth have their perceptions colored by all the from the Eric 
Raymonds and Richard Stallmans of the world. Then they wonder how 
this OSS army is going to take over the world when they can't get 
their act together and produce easily usable apps written in clear 
code. Individual developers and teams do all the time, but that's 
hardly an army.

It is amazing how we Open Source can divide people. There is gray 
all over the place and this is not a black and white issue. I think 
this is not about the armies of developers, or the quality, but about 
the ability to move fwd and construct on the disclosure of code. 
For as much bad applications there are in Open Source, there are 
good ones, OpenBSD, GCC, Postgress, and the one Rebol incorporated 
AGG.

Brian, exactly. The fact that MS still dominates the world of OS 
and Apps when there is so much OSS is a very good example.

I would never have thought that we would still be here. And it is 
disappointing to watch Europe and Asia now fall in step with USA 
on total MS adoption.

I had my son use it at college, but he had so many problems with 
it and the way it printed, that he had to stop. I've had that same 
experience with it.

So I would not say that open source model is bad or better that the 
proprietary model when producing quality applications. They are just 
models. The advantage of Open Source is that some day in the future 
anybody digging on an scavation in america will find an OpenBSD or 
Linuxe Source CD and my be able to learn from that src code. Because 
there is a mass of distribution of this CDs and it is available wideley, 
I think people will avoid reinventing the wheel. Which usually happens 
when a proprietary project go bust, and the only ramains are in tapes 
in a vault where nobody knows the *pearls* that it contatins.

And the code is so horrible and disoranized, that what should be 
done in 2 pages spans 5 modules of source code!

Thats the reason some projects flourish and others die, or lag behind. 
AGG vs Cairo.

So, it has been for me extremely rare to find it educationally useful. 
However, there are the gems, and things like AGG are shining examples. 
I wil grant you that. (But then, I do not try to understand AGG, 
I just accept it for a masterpiece that it is.)

I think the community around some of the projects is equally important, 
and also if the main person who is pushing it is willing to keep 
making it shine.

So black boxes are easier to comprehend and reproduce than others, 
but does it make sense to waste time when information can be free.

it's a load of crap. who has the resources to go through that? and 
do they really expect to find security holes by doing it?

It is very context and need sensitive.

For instance, you may ask "why reinvent the flashlight" -- but with 
the LED now that is happening, and the perfect flashlight is not 
here yet.

Best I guess is full licensing. But I believe that is expensive. 
I can not imaging licensing cost for DB2, AIX, and IBM Smalltalk.

I'm sometimes thinking about, how I should make money of a project 
like Canvas. I could make it close source and make .exe files with 
the SDK, and then maybe sell it to some niche markets. Is it possible 
to make money from it and have it open source at the same time? Any 
ideas?

Is it possible to make money from it and have it open source at the 
same time?

 Demonstrate where your solution(s) add value and then provide a service 
 (instllation, training, support, customization, etc) that folks need 
 / want to be able to use your "free" software effectively. I've said 
 this to a few people here already, but I've made quite a few dollars 
 selling OpenOffice installation and support to folks ... what's important 
 is the relationship you build with your customer(s); the fact that 
 you can under bid a proprietary solution and still make better margin 
 is just icing on the cake. ;)

Also, you have demostrated this by releasing RebDB and RebGUI. Which 
I believe are components of some of your commercial applications.

FYI: The Debian version of /view and /Core run out of the box on 
Slackware 10.2 ...

The library's from debian are pretty straghtforward..and thus many 
linux distributions should be able to run that too..If not they they 
have a problem and not rebol...

Just tried the latest View for Linux(debian build) and it works OK 
under Mepis Linux here.

you need to find a path to your fonts and you can use any ttf font. 
I don't know what YALN(yet another Linux distro) are you running 
;)

oke.. with 32 and 64 in size ?

Its strange.. I should not be needing to run a fontserver..My desktop 
is already anti-aliased with fonts and so it the rest..But rebol 
still does not display it... mmm It does load the font though..(also 
when i look with an Strace during the execute rebol does read the 
font..) it simply does not display it ..yet! ;-)

>> probe fnt1

make object! [
    name: "/usr/X11R6/lib/X11/fonts/TTF/VeraMono.ttf"
    style: none
    size: 32
    color: 0.0.0
    offset: 2x2
    space: 0x0
    align: 'center
    valign: 'center
    shadow: none
]

Well.. all applications on my linux machine use and anti-aliased 
and truetype fonts....so that config is correct.. But what i try 
or what i run on fonts Linux simlpy doesnt display the font.. It 
loads the font.. (checked that)...but thats it.. just a white field.. 
So .. I give up... ;-(

And I was wondering why there are new Linux builds when the old ones 
work without problems :)

and i LOVE fonts ;-)

The freetype support should work properly only with the new Linux 
builds: 16-Mar-2006 and up

Henrik: the FreeType suport in Linux ports was added recently and 
there was some inconsistency problems how the FreeType wrapper worked 
between different distros. Now the code should be 'united' to work 
ok.

I've tried to "rebol --reinstall" and that's funny, there's Windows 
installer with "create start menu item" and "create desktop shortcut" 
and with the great ending: 

** Script Error: create-reg has no value
** Where: register
** Near: if not create-reg/hklm key [return false]

(latest debian release)

petr: the "proper" way is to link to gtk or qt. you know, there's 
no gui standard there... the only thing you can take for granted 
on GNU/Linux is the linux kernel and the GNU utilities.

Yep, but the way Linux distributions have evolved, it's obvious it's 
impossible and one doesn't even think about it :-/

A while ago I tried to install VMware player and it nuked my system. 
Upon investigation, it had 13,000 lines of PERL script for installation 
and configuration. The major part of that was to identify the system 
parts versions and condition...

and now we have freedesktop.org and other standards organizations. 
I think basically it's like shoveling crap uphill. I wonder if it 
will get anywhere.

I think the Ubuntu guys are doing a good job. the problem is that 
there are 20 other organizations that are almost doing the same jobs. 
it's an incredible waste of time and effort.

I think we should take AROS and change its name.

which is how I decoupled the refresh and the event handler of my 
graph editor.

liquid is a hybrid of storage, association, synchronisation, dependencies, 
and some in between.  it does all of those things using the same 
core node, and allows you to build any kind of manager, or dependecy 
model over it.

like  a node which has "one" and another which has 1

yes... already part of the engine... and again... the magic is doing 
so with the same actual object class   :-)

and its all VERY stable.  I've already got a view engine running 
the exact thing you describe where a scroll bar updates a field and 
vice versa... although one is a decimal and the other a string.

HAHAHA why do you think I have been designing liquid for the last 
4 years (and actually more, in design :;-)

graphic designs piped into button images, and docs... change the 
shapes, and see your running application's bitmaps update while you 
are editing the image in the vector paint package  :-)

well, I'm happy that a few of you are starting to come out in anticipation 
(some in private  :-)  this means my platform will have appeal for 
many of us and we will be able to start a real rebolution.

regraph uses it at about 80%  and liquidator at about 95%...

don't know had a lot of issues dragging thing around... but in liquid, 
this stuff doesn't even have to be coded... and its all uniform the 
same generic api for all data... which is why it pays off in the 
end... you stop trying to design the actual dependencies, and start 
working on the algorythms.

don't I have been using it commercially and its currently very stable.

and u will nost post a first version of your interactive node editing 
tool before devcon ?

but its not yet actually able to DO anything.  I'm at the fledgling 
days of optimising the gfx engine for the graph editor.  The node 
creation, dragging and linking are done, (all graphically) and even 
already detect cycles, so you can't connect nodes in such a way as 
to get a dependency cycle  :-)

cause I'm basically hacking away at a working version and breaking 
it completely, hoping to patch it up and get it working with more 
features and cleaned up afterwards.

what we can do is manage code and edit the code on the fly... which 
is what I will be doing as my first liquidator module.

liquidator's first module I wish to build is  a REBOL IDE.  so that 
means...  a decent editor, code compositing, concurrent souce version 
control... right on the source without CVS (no need), notes, unit 
testing, etc... and debugging, when possible.

and slow

yes but to achieve this project and doing a real emulator, i need 
rebcode

Maxim: I know at one point you were looking at Sentences and the 
associative model. This sounds similiar: http://www.pilesys.com/new/news.php

so far its exceeding my expectations.  although not dialected, the 
engine itself is very pliable and the lazy computing seems to pay 
off in general.

liquidator is a good test bed for the engine, and so far, I have 
changed nothing in the design of liquid itself, I just keep improving 
how I link stuff and manage the liquid nodes themselves.

also, the fact that symmetric piping and dependency trees can co-exist 
with such triviality makes integrating GUI within networks really 
easy so far.

so... its a bit tough at first, cause you can't just go in and start 
off quickly... especially since some coding practices have to change 
to adapt to the paradigm... but in the end, you have no cleanup phase. 
 so its a fair tradeoff...

I implemented a simple session login with a neophyte on this list... 
and it was a good learning experience for both  :-)

it just turns programming upside down and you have to think so differently 
that some impossible things are just plain trivial, and stupidly 
easy things become a design nightmare!

also since we have explicit knowledge of dirtyness of data, we can 
block I/O explicitely before or after some inputs have processed... 
 so if you have  a 'hide state for example, actually changing that 
state can send (or not)  an update, so that the other inputs get 
used when it un blocks... and any one needing the value, will get 
the modified values, which where stored while the node was blocked 
 :-)  no data is lost, its only dormant.

applying this to a gui driven with liquid nodes, you could freeze 
the the whole layout on a modal window... and let your inputs continue 
to process in the background... updating animation, and reacting 
to async reads... for exacmple.


when you unfreeze the gui and call a refresh of the gui plug, all 
the data which was being processed in the background, is now automatically 
available ,as if nothing had been frozen and a simple update of the 
node, will refresh you gui with nothing to manage.

Yes, looks good.  


May I suggest doing a "bright version" (white BG...etc.  As if Mac 
designed this...)


Although the dramatic colours always look striking, in the long run 
people tend to prefer the light colours so that they can use them 
easier in print and slide shows.

and in fact, using elixir, many of the traditional need for "standard 
comms" becomes irrelevant as you build up apps on the fly, exactly 
as you need them.

the cool thing is the cross application capabilities.... need to 
insert a pic in a button... hum.. just use the output of the image 
creating node and plug it in the image input of a button.

I know... and the gobs, match conceptually my globs... funny how 
the naming even got to be almost the same.

so if its done Like I think, using R3 elixir should be VERY interactive. 
 and if that doesn't work, well, I can just use OpenGL later on.

I am curious, what Carl has in mind for View ... there were view 
plug-ins planned (dunno if it is the same concept as "rebol language 
plug-ins"?), which were supposed to provide you access to loaders, 
savers, and maybe even buffers?

its what I have read between the lines... only the core is said to 
be closed source.  and much of the reason for going open source IS 
view... many bugs would have been fixed ages ago.. I think Carl sees 
the advantage.

and I think he also understand the need for some of us to NOT be 
standard REBOL cause we are not writting small scriptlets.

If I decided to go ahead with OpenGL and Elixir, using GLASS as the 
design... what is left of REBOL but the syntax and the interpreter 
anyways?  I'd rather have the tool running on steroids without being 
tied to a chain.

python uses tcl and litterally converts python calls into tcp script 
and applies them as strings through the tcp interpreter... its soooo 
lame.

same kind of list-based, free-form datatype and intutive.