from:"QuantumG"



Speaking of static analysis.  I managed to get an evaluation version of 
this software:


   http://www.eschertech.com/products/

It works by changing the level of representation at which you write 
software.  Instead of writing imperative programs you write declarative 
specifications.  The problem is it's:
  
   1. a new, unfamiliar and difficult to master syntax (and way of 
thinking).

   2. useless for already developed software.
   3. limited in obscure and unobvious ways.

In particular, you can write a full and complete specification, ask it 
to generate a program which implements that specification and it will 
barf.  You've done everything that should be required but the state of 
the art of specification refinement that Perfect Developer embodies is 
not sufficient to generate code for any given specification.  So you 
have to write code to do the hard bits of the specification, stating 
how the program should operate, not just what the program should do.  
That'd be fine if you could do it in a real programming language, but 
instead you've gotta use the specification language.  Programming in a 
specification language is much like chinese water torture.


But it gives you an idea of just how extreme static analysis can be.  
You literally can write some specifications in Perfect Developer that 
can be verified for internal consistency, automatically refined into a 
working prototype with good performance and deployed as is.  Maybe in 10 
years time all critical systems will be written with tools like this?  
Maybe in 20 years time all other software will be written with tools 
like this?


Trent

--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Re: dynamic vs static type checking


Benno wrote:


There is also Isabelle: http://isabelle.in.tum.de/ which is free.
 



I've seen this before but run away screaming because it is Emacs based.  
Two options, I can try to learn Emacs (shudder) or I can try to figure 
out how this thing is integrated with Emacs and use it without Emacs.


I think the former would be easier, but I'd feel so dirty afterwards.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Your top-ten linux desktop apps


Erik de Castro Lopo wrote:


Nice troll or was it?
 



read The End Of History And The Last Programming Language.

Best I can find for a web reference:
  
   
http://www.cs.iastate.edu/~leavens/ComS541Fall97/hw-pages/history/gabriel.html


Basically if your language is new and you don't have a C syntax, 
you're unnecessarily placing yourself outside of the mainstream.  It's a 
bitch, but it's just syntax, so why not make it intuitive to the vast 
majority of programmers (however unfortunate that may be).


Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Your top-ten linux desktop apps


Erik de Castro Lopo wrote:


All you've done is replace the for loop with a while loop. You are
still setting the start condition and the end condition for the 
looping operation. These are things the compiler (or rather the

language) expect you to do.
 



In Io (a dynamically typed language) you'd do:

   array := List clone append(3, 2, -1, 4, 11, 1231232)
   array foreach(i, v, writeln(array[, i, ] = , v))

and foreach tends to be defined for just about every data type so you 
get a lot of consistency.


But yes, you're right, the C language should have a foreach keyword.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Installing X in Debian


Peter Hardy wrote:


Then you need an X server (the program responsible for accepting input
and actually drawing on the screen), 



Which reminds me.  Does anyone know the status of Gnome/DirectFB?  Last 
time I checked they were finding DirectFB a bit of a moving target.


Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Your top-ten linux desktop apps

2005-09-26 Thread QuantumG


Erik de Castro Lopo wrote:


Thats why I'm so keen on O'Caml. It offers even more static analysis
than C and C++. Its significantly more difficult to write bugs into
an O'Caml program than a C or C++ program.
 



Sounds like the antithesis of Objective-C and other dynamically typed 
languages.


Fun.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Your top-ten linux desktop apps

2005-09-26 Thread QuantumG


Erik de Castro Lopo wrote:


You will notice that something like the Array.mapi function is
much less likely to contain errors than the C for loop.
 



What I noticed is that they invented syntax when they could have just as 
easily have used C syntax.  Way to knife your language.


Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] request contact from computerbank

2005-09-23 Thread QuantumG


Taryn East wrote:


anyway - hopefully the right somebody will see this message and respond
offlist :)
 



I'd like to hear from them too, my partner comes into a lot of old 
(pentium II/III) systems which the company she works for just throws 
away.  She's started to bring them home but we don't have the space.


Thanks,

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Presentation Mind Control - 21st September 2005

2005-09-20 Thread QuantumG


Terry Collins wrote:


3) Unfortunately, Jacinta earnt her bile.
 



I don't think Jacinta chose the name of the talk.

I think Jeff is more deserving of bile, pile it on him.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] great code to learn from - request

2005-09-20 Thread QuantumG


Taryn East wrote:


what nobody else is going to bite? :(
 



I heard someone bitching the other day that gtk+/python apps are slow.  
Not been my experience, but if you're sufficiently bored, why don't you 
download some and see for yourself?


Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Presentation Mind Control - 21st September 2005


Jacinta Richardson wrote:


G'day everyone,

You are invited to join us in a talk by Paul Fenwick about Presentation Mind
Control - how to make other people think your talk is much better than it really
is.  Come and learn some great tips on how to improve your presentations.
 



Ok, I'll bite.  WTF.  If you have something worthwhile to say, geeks 
will listen to you.  We don't need this motivational speaking crap.  Or 
is this about making presentations to the mindless masses?


Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Presentation Mind Control - 21st September 2005


[EMAIL PROTECTED] wrote:


Does it make sense now?
 



Mindless masses, got ya.

Trent

--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Presentation Mind Control - 21st September 2005


David Kempe wrote:



I talk to plenty of geeks who have something worthwhile to say, but 
can't make the words make sense. Its not motivational speaking, its 
effective communication. Entirely different. Geeks are just people, 
people listen to effective communicators, regardless of the worth of 
their speech.



And rather than call the workshop Effective Communication For Geeks they 
call it Conference Presentation Mind Control..


Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Presentation Mind Control - 21st September 2005


Jeff Waugh wrote:


Have you ever kissed a girl?
 



You're a disgrace.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] OT: Interesting Article about 64bit computing

2005-09-12 Thread QuantumG


James Gray wrote:

Peripherals you currently own or plan to purchase in the near future 
(cameras, printers, MP3 players, handheld devices) will most likely NOT work 
on a system purchased with Windows XP Professional x64, and some software 
applications may not work on the x64 operating system.
 



And more importantly, a company will only port a driver to x64 for their 
latest and greatest product.  They're not going to port drivers for a 
web cam made 5+ years ago.  So adopters of Win64 will either have to buy 
new hardware or go without.


Trent

--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Aptitude command to TOTALLY remove KDE

2005-09-01 Thread QuantumG


Sam Couter wrote:


No, what I said is correct. The kernel is largely irrelevant to the
end-user experience.
 



Whether or not the kernel has real time scheduling or not makes one heck 
of a difference to the end-user experience.  And, (for I think, the 
third time now?) although you can hack real time scheduling into a 
monolithic kernel it's not a sane thing to do.


Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Aptitude command to TOTALLY remove KDE

2005-09-01 Thread QuantumG


Jeff Waugh wrote:


And for the third time, you haven't suggested any reasons why you think this
is the case. There is more to user experience than the extreme technical
definition of responsiveness suggested by real time scheduling.
 



Obviously.  But real time scheduling *does* contribute to that 
experience.  Which is the claim that I made.  So why are you arguing?


Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Re: Aptitude command to TOTALLY remove KDE

2005-09-01 Thread QuantumG


[EMAIL PROTECTED] wrote:


Don't forget that a microkernel introduces communication overhead and
usually some extra scheduling overhead which in turn eats into performance.
I seem to remember there was a big squabble over who had the fastest
webserver until Linux introduced a kernel level http accelerator which
blew everyone else out of the water so badly that they first tried accusing
the kernel developers of cheating and when that didn't work they just
stopped playing that game, took the ball and went home.
 



That's because it's a stupid idea.  Really, it is.  We can put all the 
code of the entire system into one process and run it in ring-0 and 
we'll have the fastest system in the world!  Until it crashes.  Which is 
what the argument really boils down to.  When something which is part of 
the kernel crashes, do you want it to take down the whole machine or 
do you want it to be contained and replacable?  A web server should run 
in user space.  End of story.  Debatably a file system and a network 
stack should run in user space.. if you can get sufficient performance.  
For a desktop operating system (which, remember, is what we were talking 
about) today's hardware is so much overkill that you could run different 
parts of the kernel on different parts of a local network and still get 
adequate performance.  So why are we still running it in ring-0?  
Because that's the system we have today and it doesn't make sense to 
reinvent the wheel if it is working.  But it does make sense to keep 
working on it and slowly migrate things out of ring-0 and into their own 
process space so we can get rock solid stability and the flexibility 
needed to implement and debug new things more safely.


Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Aptitude command to TOTALLY remove KDE

2005-08-31 Thread QuantumG


Sam Couter wrote:


The Hurd is just a kernel (well, a microkernel and a bunch of servers
that offer services normally provided by more traditional kernels).
GNOME runs on the Hurd and it's about the same as GNOME on Linux or
FreeBSD or any one of a bunch of free operating systems. I don't know if
KDE runs on the Hurd yet, but when it does, it'll be just like running
KDE on Linux or FreeBSD or any one of a bunch of free operating systems.
 



See, this is flat out wrong.  We've know since BeOS that a microkernel 
makes for a better desktop operating system.  The denial that we can 
just keep hacking support for real time systems into monolithic kernels 
is just insanity.  All the arguments made for the architecture of the 
Linux kernel, in particular that it's simpler and you don't need 
features like kernel threads and kernel re-entry, have long since been 
abandoned by the Linux community itself.  So to keep arguing at this 
point that Linux is using the best architecture for a kernel is 
redundant.  We know it's wrong.


That is not to say that we should just throw Linux away.  In fact, I 
advocate the opposite.  We need to refocus work on the Linux kernel and 
step up the incremental migration of services out of the kernel.  
There's been great strides at moving filesystems into user space 
already.  Moving networking and device drivers and, finally, memory 
management and scheduling into userspace is the next logic step.  
Unfortunately it's gunna take years.


/rant

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] MicroWHO? [Was: Aptitude command to TOTALLY remove KDE]

2005-08-31 Thread QuantumG


Jeff Waugh wrote:


Say *what*?! There were no significant features in BeOS that are impossible
to implement on Linux, either in the kernel or in userland. Attempting to
credit the vast number of variables in a high quality user experience on a
specious technicality is ridiculous.
 



I didn't say it was impossible.  I said it was insane.  Yes, you can 
hack real time scheduling into Linux.  Yes, we do have a kernel threads 
and kernel re-entry in Linux.  Does Linux have the best architecture for 
such features?  Hell no.  The result is a massive blob of complexity 
running at the highest privilege level on the system.


As for whether or not a microkernel really made a lot of difference to 
BeOS.. it contributed a lot to the overall system architecture.  It let 
the developers compartmentalize concerns into communicating servers more 
readily than a monolithic kernel does. 


Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Aptitude command to TOTALLY remove KDE

2005-08-30 Thread QuantumG


Jim Lemon wrote:

This is another of those interesting discussions that sometimes veers 
close to a Creationism vs Intelligent Design slanging match.


1) Why do Apple and Microsoft spend more than the GNP of some 
countries in legal fees fighting over user interfaces?


Once you've got your user on autopilot, change is disruptive. 



Heh, funny you mention this, I was just looking at this

   http://www.xpde.info/

They don't appear to have a Windows Explorer (the file manager that 
often crashes taking down the entire desktop) look-a-like yet.  Which 
means they won't get far with your ingrained Windows user, but it's a 
nice start.


I actually do think this is the way to go if you want to get Windows 
users to switch to a Linux system enmass.  Once they've settled into 
their new (but similar) environment they can start playing with 
different window managers and the like.  This whole cold turkey switch 
campaign that people seem to put forward for switching to Linux is a bit 
too hardcore IMHO.


Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Software Freedom Day

2005-08-24 Thread QuantumG


Pia Waugh wrote:


Yes! We are planning to hand out CDs and foo in the city followed by an
installfest. Details currently at
http://maitri.ubuntu.com/softwarefreedomday/wiki/index.php/Sydney and are
yet to be finalised. We'd love more hands to help out if you can on the day!
I have tshirts on the way so we all look good :)
 



Cool!  Do we have a big SOFTWARE FREEDOM DAY banner and permission to 
hang it in a public place.  That'd rock.


You can count me in, what time are we starting?

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

[SLUG] Software Freedom Day

2005-08-17 Thread QuantumG



So is anyone planning activities for Software Freedom Day?

Need any help?

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Google Earth on Linux ?

2005-07-21 Thread QuantumG


[EMAIL PROTECTED] wrote:


is it possible to run Google Earth on Linux ?
 



Well, there's no linux version available from their web site, if that's 
the question.


However, looking at the 38 dlls it uses, I really can't tell you why.  
For example, the user interface, is using Qt.  To fetch data from the 
web, they're using libcurl.  To parse xml, they're using libexpat.  They 
support both directx and opengl rendering.  So yeah, gotta wonder, with 
a lot of work you'd expect WINE to be able to run it.  You shouldn't 
even need Cedega thanks to the opengl support.  Seems they could release 
a linux version, maybe you could just ask them nicely? :)


Trent

--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Is Linux for losers?

2005-06-22 Thread QuantumG


Carlo Sogono wrote:


PS. I'm not an OpenBSD guy.



Let's just say that the line to punch Theo in the head forms to the 
left.  As for Simon Lok, he's obviously never written a serious piece of 
software in his life.


Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Is Linux for losers?

2005-06-22 Thread QuantumG


Nathan Clark wrote:

He fails to mention that all code that has ever been written and is going to 
be written sucks. 



Way to have hope for the future dude.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] XP access to ext3 linux partition

2005-06-01 Thread QuantumG


linley caetan wrote:


I have just installed ubuntu as dual boot on my sony
vaio.
Is there a straighforward way of accessing the linux
patition from windows ?
 



I recently installed an ext3 device driver for windows xp.  It's way too 
unstable for actual use.  I think the simple answer is no.


Trent

--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

[SLUG] Windows XP SP2

2005-05-27 Thread QuantumG



Someone I know, who will remain names, just told me that they have just 
discovered that the latest version of their software that they have 
deployed to a few 10s of thousands of sites doesn't work in Windows XP 
SP2.  I said I wasn't surprised as, you know, SP2 is a train wreck.  
They stress to me how screwed they were as Microsoft is forcing SP2 on 
everyone running Windows Update so they have to tell their customers not 
to run it (which looks really bad).  But then the kicker happened:


   So when does the next service pack for XP come out?
   Heh, there probably won't be one.
   Huh, what do you mean?
   Well, Longhorn is supposed to be coming out this year.
   Longhorn?  What's that?

Mwahahaha.. Said person went into a small fit when I told them that 
Longhorn (or Longhaul as I pronounce it) is the next revolutionary 
operating system from Microsoft and yeah, if you're having trouble with 
SP2 you aint seen nothin' yet.


What's really stupid is that this company has tried to reduce costs by 
installing software onto client computers instead of deploying 
preconfigured computers.  If they'd just made their software for linux 
and distributed preinstalled computers to every customer they wouldn't 
have so many headaches.


Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] fastest *.zip file locator

2005-05-27 Thread QuantumG


Del wrote:



if [ -any zip files- ]; then
  do this
fi


blah=*.zip
if [ -n $blah ]; then
 echo foo
fi

is what I'd do.

TRent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] -Wshadow


Erik de Castro Lopo wrote:


OK, you are suggesting that the compiler should treat parameter
names and function names as being in separate name spaces.

If you then have a function named foo and parameter named foo
and you use foo in your function? How does the compiler 
distinguish between foo the parameter and foo the function? 

C++ has the scope resolution operator ::, but C has no such 
mechanism.
 



Use the extern keyword:

int foo() {
 return 5;
}

void bar(int foo)
{
 extern int foo();
 foo();
}

enjoy :)

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] -Wshadow


Erik de Castro Lopo wrote:


Sorry, I don't think that is an answer. In particular, it does
not match C++'s scope resolution operator.

What we need to be able to do is:

   void bar(int foo)
   {
  /* Use foo the function. */
   x = foo();

   /* Use foo the integer. */
   x += foo;

  /* Use foo the function. */
   x = foo();
   }
 



Use a different language.

Trent

--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] -Wshadow


Robert Collins wrote:


extern int foo (void);

static void 
bar (int foo)

{
 /* use foo the function */
 int x = foo ();
 /* use foo the function */
 /* (OH shit) */
 int (*y) (void) = foo;
 /* use both ;) */
 x = foo + y ();
}
 



Almost!  What you want is:

static void
bar(int foo)
{
   int x = foo;

   int (*y) (void);
   {
   extern int foo(void);
   y = foo;
   }
   x = foo + y();
}


i.e., assign the tempory y to be the extern definition of foo.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] -Wshadow


QuantumG wrote:



   x = foo + y();



oops, that should be:

   x = foo + (*y)();

which admitably is painful.  Alternatively you can make a tempory to 
hold the parameter..


Of course, this is all silly and you shouldn't do it in normal code.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] system binaries attempt to exec on login ?

2005-05-24 Thread QuantumG


Voytek Eymont wrote:


what have I done... ? they are they supposed to be executing on login... ?
 



check your startup files.. (like .bash_profile) there's probably some 
attempt to set up aliases and they expect the executables to be world 
readable.  (which is bad but not unheard of).


Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] getting all .bash_histories across the hd ?

2005-05-22 Thread QuantumG


Voytek wrote:


what the easiest way to get to all '.bash_history' files on the whole drive ?
 



locate .bash_history

will probably do it, if not:

find / -name .bash_history

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] setting system binaries to 0700 ?

2005-05-22 Thread QuantumG


Voytek wrote:


ooops I guess I meant execute and read, not write...
 



Why do you need to read it?

Trent

--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

[SLUG] glut apps giving blank window

2005-05-18 Thread QuantumG

Would anyone happen to know why an app which uses libglut might display 
a blank screen under ubuntu but work fine on every other linux box I've 
tried?  I'm thinking xorg might be the issue.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Clipboards under X

2005-05-17 Thread QuantumG

David Gillies wrote:
Now under ubuntu selecting text and ctrl+c text appears to be two
seperate clipboards. This can be quite annoying since some applications
can't ctrl+v to paste and others I have can't middle mouse click to paste.
 

hehe, if you think that's annoying, ctrl-c from an app, close the app 
and then ctrl-v somewhere.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Compatibility Issues with Windows64

2005-05-05 Thread QuantumG

Edwin Humphries wrote:
 Federal Government bodies are concerned that there are Linux 
compatibility issues with upcoming 64-bit Windows.

Uhhh, I'd personally be more worried about the absolutely certain 
incompatabilities that exist between win64 and winXP/2k.  It's so hard 
to believe that the world is going to willingly switch to win64 when the 
vast majority of 3rd party hardware currently in existance doesn't work 
with it.  In many ways, it's like Linux was like back in the 90s.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] gunzip error: invalid compressed data--format violated

2005-05-04 Thread QuantumG

Michael Lake wrote:
Thats why I'm wondering if it's an error because I used an IRIX gzip 
and trying to unzip with gunzip on a TiPowerBook under Debian.

I have had similar issues.  I actually installed gzip from source on the 
ancient machine I was using.  These days I'd install bzip2.

Of course, maybe you can locate some source for this ancient IRIX gzip.
Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

[SLUG] ext2 ifs driver

2005-04-26 Thread QuantumG

At the Ubuntu conf yesterday a number of people brought up the issue of 
transfering files from linux to windows on a dual boot machine.  
Apparently the vast majority of people still use a FAT32 partition to do 
this (or a FAT32 formated removable drive like a USB stick).  I just 
thought I'd point out to those on this list an alternative.  No-one 
really worries about putting stuff onto FAT32 in windows to get from 
linux because linux can read NTFS partitions reliably.  If windows could 
read ext2/ext3 partitions no-one would worry about putting stuff onto 
FAT32 in linux either.  Well, it can be done!

   http://uranus.it.swin.edu.au/~jn/linux/ext2ifs.htm
I managed get this working, although it really could be a lot easier.  I 
downloaded ext2ifs-0.3.zip 
http://uranus.it.swin.edu.au/%7Ejn/linux/ext2ifs/ext2ifs-0.3.zip but 
discovered that there's some bug with the installer program.  The author 
has provided a fix for this on the same page, service-0.2b.exe 
http://uranus.it.swin.edu.au/%7Ejn/linux/ext2ifs/service-0.2b.exe 
which I downloaded.  It required some more fiddling, to actually see the 
buttons I had to press (which were anything but obvious).  Step 1 was to 
install the driver.  Step 2 was to mount the linux partition as a 
drive letter.  This didn't work.  There's 3 different ways to do it and 
none of them appeared to work.  Step 3 was to start the driver, which 
although it appeared to work, didn't actually result in anything useful 
due to the failure of step 2.

I actually had to reboot to linux to get some files anyway, so I did 
that, using a usb stick to transfer them to windows.  Upon rebooting I 
went to My Computer and discovered that my usb stick was no longer E:, 
it was a harddrive partition.  Well that's interesting, so I double 
clicked on it and Microsoft kindly told me that this drive had not 
been formated and would I like to format it now!  I very, very carefully 
clicked NO and then fired up the server-0.2b.exe file and performed step 
3.  Double clicking on the E: icon now gave me access to my linux partition.

So although I'm sure the average user could quite easily trash their 
computer with ext2ifs as it currently exists, I'm confident that with 
some polish it could be installed safely, with sensible defaults, such 
that a beginner would have no trouble accessing their linux partition.  
For a default Ubuntu installation which, I believe, puts /home on its 
own partition, this is probably the ideal partition to make available 
under windows.

Hope that was helpful to someone,
Trent Waddington
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Can anyone tell me what ns1.aztec.20.com is???

2005-04-25 Thread QuantumG

Alan L Tyree wrote:
CRIMINAL CODE ACT 1995
- SECT 476.2 Meaning of unauthorised access, modification or impairment 

(1)
   In this Part: 

(a) access to data held in a computer; or
(b) modification of data held in a computer; or
(c) the impairment of electronic communication to or from a computer; or
(d) the impairment of the reliability, security or operation of any data
held on a computer disk, credit card or other device used to store data
by electronic means;
by a person is unauthorised if the person is not entitled to cause that
access, modification or impairment. 

WTF does not entitled mean?  Does it mean you have to have written 
permission to access a computer.  In which case how is it legal to 
connect to a web site?

None of this stuff has been thought through or tested in court has it?
Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] first day of week in Ubuntu Hoary?

2005-04-10 Thread QuantumG

Sonia Hamilton wrote:
I've just upgraded one of my machines to Ubuntu Hoary - very nice.
I've noticed however that the first day of the week defaults to Sunday
rather than Monday, and I can't find a way of changing it. Any pointers?
 

The religous implications are astonishing!
Seriously though, I looked into it and there is a gconf variable for 
ecal called week_start_day but I don't think it is related/honoured by 
the date applet.

I've just upgraded to Hoary too :)  I actually managed to get my nvidia 
TwinView working on the second monitor.. although I did have to hack my 
xorg.conf file myself.  Perhaps I should try hacking up a nice Python 
app to configure monitor layout?

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] [ot] Rant on tech salesman who has not heard of Linux

2005-04-08 Thread QuantumG

Richard Hayes wrote:
Hardly Normal is the largest retailer of computer equipment is 
Australia by a long way.  If their salespeople have not heard of Linux 
then I fear Australia is further behind that I thought.

I got a lecture yesterday from the uber-nerd at EB when I asked him when 
they were getting in The Matrix Online.  His first question was which 
country? to which I replied, after a significantly long pause, this 
country.  He went on to say that unless a game has a lot of money 
behind it like Blizzard or Sony then it's not considered economical to 
release it in Australia.  Apparently Warner Bros isn't a big company to 
EB staff.

The relationship to Linux?  He was treating me like someone who knew 
nothing about computers even though I clearly had a Linux T-shirt on.

More startling: I went home to buy the game online, only to discover 
that Amazon doesn't ship *any* video games outside the US.  I eventually 
got myself a copy from GameSpot.com.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] problem with my ubuntu

2005-04-08 Thread QuantumG

ray wrote:
hey there all i have a major problem with my
Ubuntu platform it won't start up anymore and tells me the X sever has 
inexplicable errors and that i need to reset it to get the GDM working 
properly .Any how its all too complicated for my meager abilities and 
I can't start the thing to get on line and find the  answers ( i'm 
sending this from my brotheres computer ) My mate in Coffs said it 
sounds like a harddrive error and I've lost data but  he's so far away 
and not sure and it costs a fortune to call ,I'm at a loss what to do 
any suggestions ,RAY 

The basic process is to look at your  /var/log/Xfree86.0.log and 
determine what isn't working.  Then you modify your 
/etc/X11/XF86Config-4 to fix it.  But you'll probably need someone to 
walk you through it.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] excrypting fs

2005-04-07 Thread QuantumG

Howard Lowndes wrote:
Probably a stronger front door if the boys in blue come knocking with 
heavy hammers because they suspect that you have something to hide :)

All the cool kids these days run Knoppix and just power down their 
machine when their Mum^H^H^H^H^H police knock on their door.  Poof goes 
the ram disk.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Virtual hosting with VMWare

2005-04-05 Thread QuantumG

Broun, Bevan wrote:
I think you will want vmware esx which installs straight on your hardware.
I used to work at VMWare.  All the new features (especially performance 
features) went into ESX before they went into GSX and sometimes filtered 
down to Workstation.  It's a little known fact that VMWare actually 
writes their own kernel, specifically for running virtual machines.  I 
remember that page sharing and fair scheduling was the two major 
advantages of ESX.  Don't know how much has changed since back then.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Virtual hosting with VMWare

2005-04-05 Thread QuantumG

Broun, Bevan wrote:
Can you clarify this for me. is vmware esx it's own kernel or is it linux
with vmware's custom modules (which is how it' appears to me)?
 

IIRC, GSX is Linux with custom kernel modules but ESX is it's own 
kernel.  I think it was based on FreeBSD way back when they founded 
VMWare.. no idea if they pull things from the FreeBSD of today.

I havn't worked there since 2001 and my memory is a bit foggy :)
Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] wget *.jpg

2005-04-03 Thread QuantumG

Voytek wrote:
Is there a way with wget to get just all jpegs of a web server tree ?
I'm reading the docs, but haven't figured it out yet
 

it's something like:
  
   wget -r -A *.jpg www.myserver.tld

I did something similar the other day.
Trent

--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Nomination acceptance

2005-03-20 Thread QuantumG

Grant Parnell wrote:
* I want to get more youth involved because a) they have more time to do 
stuff and b) I like teaching stuff.
 

Can I just mention another good reason to encourage kids who are 
interested in programming to start hacking on open source projects:

   kids have a better understanding of fun
Maybe I'm just cynical but it always seems that so many hackers these 
days see working on open source as work.  Maybe if we had more kids 
hacking on open source projects they could remind us of all the joy we 
get out of it.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] proftpd

2005-03-15 Thread QuantumG

Luke Skywalker wrote:

drw-r--r--4 exangel ftp 128 Mar 16 09:43 ftp_private

Well I don't know about the rest of it but presumably you want +x on 
that dir, otherwise no-one will be allowed to cd into it.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Thunderbird as default e-mail for Firefox - not succesful yet

2005-03-13 Thread QuantumG

Peter Rundle wrote:
My only gripe is that prior to this, in mozilla 1.6... you could right 
click on a link in the e-mail browser and open the page in a new tab, 
where as thunderbird doesn't appear to support this, it opens it in 
the active tab replacing the page of interest you might have open (I 
guess it's trying to be browser neutral).

In FireFox, under Options - Advanced - Tabbed Browsing there is the 
option Open links from other applications in: and you can select a 
new tab in the most recent window.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] How do I mount an audio Cd ?

2005-03-12 Thread QuantumG

Jeff Waugh wrote:
Windows is lying a little bit, to give you a nicer interface. Audio CDs are
not like data CDs, and cannot be mounted.
 

From a purely philosophical point of view, what would be a good reason 
for not have a kernel module that mounts audio CDs by interpreting the 
red book format?  Seems kind of silly to have code at the application 
level doing this low level interpretation.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Using open source code in propriety code

2005-03-09 Thread QuantumG

Richard Hayes wrote:
Dear List,
What OSI approved licenses allow for the reincorporation of open 
source code into propriety products.

BSD obviously.  BSD not only allows for it but actively encourages it.
Any others?

I believe the term for this is non-copyleft.  But that actually can be 
a little deceiving.  The LGPL is a copyleft license, but practically, 
most proprietary software developers have no problem using software 
under this license as they are free to dynamically link to it.  They are 
required to make available any modifications they make to software under 
an LGPL license, but this usually isn't a problem as its not the 
modifications they are trying to keep proprietary. 

   http://www.gnu.org/licenses/license-list.html
is a well maintained list of Free Software licenses.  Any that say 
non-copyleft are likely to be compatible with creating proprietary 
products.

Now, I really do have to ask, why does your software need to be 
proprietary?  If you're trying to keep something secret by not supplying 
the source code to your program, I recommend that you seek out a 
software protection company to supply you with some hardening of your 
binary, as compilation != encryption.  If you're simply trying to keep 
control over the distribution of your software, I'd recommend you 
consider providing source code under the same restrictive license as 
which you provide your proprietary software. 

At least then your customers will have some of the advantages of Open 
Source, even if it is in a look-but-don't-touch style.  Apart from 
giving your software a unique advantage over other proprietary 
competitors, it would also ease the transition from a closed development 
model to an open one should your customers ever wise up and begin to 
value their right to distribute their modifications.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Using open source code in propriety code

2005-03-09 Thread QuantumG

[EMAIL PROTECTED] wrote:
Like LGPL?  That's different of course.
But what other copyleft licenses are there?
 

There's about 10 on this page:
  
   http://www.gnu.org/licenses/license-list.html

But the biggest one in use that you might not have considered is the 
Apple Public Source License.  In some ways it's even more restrictive 
than the GPL in that externally deploy is poorly defined.  For 
example, if you use your Mac to serve web pages and you have happened to 
modify some APSL licensed code so your server serves pages a bit faster 
you're required to make those modifications available because you've 
externally deployed them.  Think they tried to fix this in version 2.0 
of the license but it's still not clear.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Memory-mapping source files for faster lexical analysis

2005-03-01 Thread QuantumG

Michael Knight wrote:
Hi all,
I'm starting an honours year at uni this year. My supervisor and I are
thinking of doing a paper on trying to determine if memory-mapping
source files will speed up the lexical analysis phase of various
compilers (instead of whatever buffering method it currently employs).
 

Wooo, a fellow compiler geek.  Cool.  I have no heard of any papers on 
the subject you mention, no.  May I ask who your supervisor is (and at 
what university?)

If you're interested in unique uses for compiler technology, you might 
my open source project interesting:   

   http://boomerang.sourceforge.net/
We do it all backwards :)
Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Re: Weird login behaviour

2005-02-28 Thread QuantumG

Andrew Bennetts wrote:
You claimed:
o one can load kernel codes other than loadable kernel modules(LKM) that are
'enabled'.  If you try to load an LKM that is not configured the Kernel will
not allow it. And because only a dozen or so LKMs are enabled instead of,
perhaps, hundreds LKMs, it is easy to manage these.
   

This is false, but you still haven't admitted that it was anywhere that I
can see.
I dunno if anyone else has said this or not, but /dev/kmem and the joy 
of kernel exploits can allow an attacker to taint a kernel in ways that 
you simply cannot detect.  Not to mention the fact that kernel modules 
are not signed or checksumed in any way so it is trivial to modify them 
on the disk so they are tainted on next reboot. 

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Re: Weird login behaviour

2005-02-28 Thread QuantumG

[EMAIL PROTECTED] wrote:
But your message made me wonder - is it practical to disable creation of
/dev/kmem?
 

Sure is.
What other practical ways have we got to avoid attacks through it?
SE Linux? GRSecurity?
 

Well, the year I gave up being a security consultant was the same year 
that I had a frank discussion with some average hacker/cracker types 
over a beer.  I was living in Melbourne at the time and these lads all 
told me in no uncertain terms that I was a fossil.  These guys told me 
something that made me rethink the whole concept of unix security: go 
for the kernel exploit.  Every one of them had their own personal, 
never-announced-to-the-public bug in the linux kernel that they would 
use to run arbitary code.

The C language is just so full of pitfalls and traps that it is next to 
impossible to write secure code in it.  You can do it obviously (look at 
OpenBSD) but it requires so much discipline and arcane knowledge that 
most programmers simply can't write secure code.

The number of people reporting exploitable bugs in software has dropped 
in the last 5 years.  It's not because the software has gotten better.  
It's because the kind of people who used to see finding these bugs as a 
challenge that would earn them respect and admiration have been vilified 
by the security companies that see finding these bugs as something their 
employees should be doing (to earn their company respect and 
admiration).  These people didn't just stop looking for these bugs, they 
just stopped reporting them in official forums.  Instead, they sit on 
irc and tell the bad guys all the bugs they find.  Of course, there 
are the kind of people (like me) who just found the shear number of bugs 
available to be found as a turn off to the whole endevour.  I mean, how 
fun is it to search for something you know you're going to find?

So unfortunately, I have no doubt that even your average script kiddie 
has their own personal kernel exploit that will likely never be fixed.  
As my signature on Slashdot says:

   /Do we smell nothing of the security decomposition? Security is 
dead. Security remains dead and we have killed it.

Trent
/
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Stupidest law of the year candidate!

Benno wrote:
Of course, there is a bit of a problem here, how can they check it? Since that
is also illegal. 

Woah!  Good point.  I hear there's a lot of child porn over at 
207.46.144.222, let's ban that.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Stupidest law of the year candidate!

Luke Skywalker wrote:
I personally would love to see all offenders and people running those 
sites in jail...but I'd focus on taking the sites down for good
not just a restriction on access.

Well, I think the problem is that many of these sites are overseas so 
the Australian police can't touch them.  Of course, it really makes you 
wonder why there isn't international treaties on this stuff.  For 
example, I can imagine that Australia would be in favour of a treaty 
such that police officials could legally hack into and take down sites 
like this.  Of course, that's just a major slippery slope.  I'm sure it 
would be no-time before the copyright lobbists were calling for police 
to use these powers to stop infringers.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Stupidest law of the year candidate!

Howard Lowndes wrote:
Which then begs the question as to the status of M$ networks.
 

So we should ban 207.46.156.188?  Ok, doing it right now.
Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Stupidest law of the year candidate!

Michael Lake wrote:
...and what if the 15yr old model is entirely computer generated? I 
can see that in the future the business ppl that cater to this 
'market' will start to use very advanced computer graphics to create 
'kiddies'. I can see them trying to get a ruling in a court that this 
is 'acceptable'.

There's already been a ruling on this in the US.  The court found it was 
unconstitutional to censor it.  Obviously here in Australia it would be 
banned, you can't even have above age naked CG people in video games.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] list of distros

2005-02-21 Thread QuantumG

Heracles wrote:
Support for newer canon printers is problematic as canon wont allow 
open sourcing of their drivers and refuse to supply drivers for linux

I decompiled a bother printer driver for linux, it was an absolute 
mess.  Buffer overflows, hard coded values, code duplication.  They 
didn't even turn on optimization in gcc (which thankfully made it that 
much easier to decompile).

I'd be interested in having a look at the canon printer drivers.  I've 
never really looked into printer drivers on win32.  Are they just .sys 
files or what?

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] list of distros

2005-02-17 Thread QuantumG

Ken Wilson wrote:
Xandros was able to recognise all other ditributions on my hard discs
and give them a place on the boot selection screen.
 

I hear that Xandros doesn't actually try to get their patches 
intergrated back into the community projects they extend, is that true?

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] list of distros

2005-02-17 Thread QuantumG

Dean Hamstead wrote:
sorry, i just didnt like the tone of the email. im 100% debian and
bsd

Wow, I'm being chastised for things I *didn't* say now.  Cool.
Yes, as I understand it Xandros doesn't not participate in the open 
source community.  They just leech what they can get, add value and 
try to sell it.  No, there's no *requirement* that they participate in 
the community.  I just wanted to know if I had this right.  It's really 
hard to tell if Xandros gives anything back and I was wondering if 
anyone knew if they do.

Obviously when choosing to pay for a linux distribution I'd like to know 
things like this so I can factor it into my decision making process.  If 
there is a choice between two equally good, equally priced linux 
distributions and I know that one of them actually contributes something 
to the community and the other is just a leech I'd probably go with the 
one that contributes something to the community.  But that wasn't the 
point of my inquiry, I just wanted to know if I had my facts straight or 
not.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] list of distros

2005-02-17 Thread QuantumG

Rod Butcher wrote:
I think in a capitalist system, all these leeches are in fact
contributing something, 
they're filling the user-friendliness niche,  i.e. they've turned Linux
from a geek hobby into a system useable by the techophobe public.
 

Oh yes, they're contributing something to their customers, but they're 
not actually contributing anything to our community.

Then again, I guess you could say they're reducing the network effects 
of proprietary software, making open source more acceptable, that's 
definitely something.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] C newbie seeks directions

2005-02-16 Thread QuantumG

Rod Butcher wrote:
Recommended newbie-friendly C mailing lists ?
 

If you have the time I'd recommend #c on undernet.
Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Wall-to-wall Windows

2005-02-16 Thread QuantumG

David Guest wrote:
Really, someone should take the IT section of SMH to task over this 
sort of crap.
---
Although the project is mostly completed, data structures will continue 
to be modified and the last vestiges of open source will be eradicated 
in coming months.
---

That's beautiful reporting.  Maybe we need to nominate Rob O'Neill for 
one of those sardonic annual awards.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Wall-to-wall Windows

2005-02-16 Thread QuantumG

Jeff Waugh wrote:
Ah, c'mon,
 

Do you sell Positive Thinking tapes as a sideline or what? :)
Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] FYI: Campaign of Mis-Information

2005-02-15 Thread QuantumG

Anand Kumria wrote:
And the downfall of the music industry is largely their own doing
anyway.
 

Well ya know, if there is to be a downfall to the music industry one 
would hope it would be as a result of their standard operating procedure 
of putting hype ahead of artistic merit.  But hey, we've been waiting, 
what, 8 years, for the Napster and mp3.com and John Perry Barlows of the 
world to take us into a new era of non-hype based music.  What did we 
get instead?  The global karaoke contest that is Idol. 

Oh well,
Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] safe(ish) single-login from website

2005-02-15 Thread QuantumG

Rob Sharp wrote:
:-$
(I'll get me coat)
 

heh, I remember I went to an interview with Yahoo, London and for some 
reason they were asking me lots of OO Perl questions.  I also had a 
flu.  After 20 minutes of saying my recollection of that aspect of Perl 
isn't too good, I finally got up and declared I'll get my coat.  
Thanks for the pleasant memory :)

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] FYI: Campaign of Mis-Information

Craige McWhirter wrote:
http://www.smh.com.au/news/National/Copyright-laws-under-review/2005/02/14/1108229936338.html?oneclick=true
http://www.smh.com.au/news/TV--Radio/Brave-new-world-for-a-couch-potato/2005/02/14/1108229927903.html
 

Anyone able to share a login for this stuff?  Stupid New York Times 
wanna-bes.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] FYI: Campaign of Mis-Information

Benno wrote:
www.bugmenot.com is great for that kind of thing.
 

username: bug_menot
password: bugmenot
for anyone who wants it.
Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] FYI: Campaign of Mis-Information

Jeff Waugh wrote:
Summary of the couch potato article, by paragraph (media analysis is so much
fun). I'm yet to find anything seriously objectionable. The article actually
summarises our point of view on copyright issues very well, and right up
front, which is unusual.
 

Yes, this article is a lot better than the first one.  I've emailed 
Charles Britton from the ACA to ask him if he actually did pledge his 
support for a two year jail sentence for bypassing the trivial 
encryption found on consumer media, as the SMH article claims.  If I 
get a reply I'll post it on the list.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] FYI: Campaign of Mis-Information

I've heard back from Charles Britton of the ACA, that first article is
definitely engaging in some old school propaganda techniques. It
contains this text:

---
Tough new laws will be introduced by 2007 as part of the US-Australia
free-trade agreement. Anyone found guilty of unlocking an encrypted
code that prevents multiple copies of a CD being made faces a two-year
jail sentence.

The Australian Consumers' Association welcomed the move but a
spokesman, Charles Britton, said: We feel that if you pay to get access
to something then you should be able to go back to it again. We would
not want to see a fair-use right given by law only to be taken away by
technology.
---

which implies that the ACA welcomes the tough new laws when actually
the ACA has only shown support for reform to copyright law to permit
format changing and off-air recording.

For reference, here's my email to Charles Britton and his reply.
==
Hi Trent
Not a misquote, but missing context - the proposal we welcome is what
Ruddock was talking about - an inquiry into consumer copying rights,
sympathetic to format changing off-air recording etc. Certainly NOT
happy about criminalisation of consumers as delivered by the recent
FTA.
Best regards
Charles Britton

QuantumG [EMAIL PROTECTED] 02/15 2:54 pm

An online article in the Sydney Morning Herald,

http://www.smh.com.au/news/National/Copyright-laws-under-review/2005/02/14/1108229936338.html?oneclick=true

has this quote:
Tough new laws will be introduced by 2007 as part of the
US-Australia free-trade agreement. Anyone found guilty of unlocking an

encrypted code that prevents multiple copies of a CD being made faces a
two-year jail sentence.
The Australian Consumers' Association welcomed the move but a
spokesman, Charles Britton, said: We feel that if you pay to get
access
to something then you should be able to go back to it again. We would
not want to see a fair-use right given by law only to be taken away by

technology.
I was wondering if that was actually correct? Did you say that the ACA
supports a two-year jail sentence for bypassing the trivial
encryption
that is to be found on consumer media? Is this the official position
of
the ACA?

Seems a bit silly doesn't it? Gee, what are you in for? Oh, I
knocked over a petrol station, got 2 years, you? Yeah, I got two
years
too... for copying a CD.

Trent Waddington

--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] FYI: Campaign of Mis-Information

Jeff Waugh wrote:
Missing context for a quote can't be anything other than an old school
propaganda technique? Sheesh... Take off your tin foil hats, guys.
 

Umm no.  Did you read the article?  The writer has deliberately gone to 
the effort of placing those two paragraphs together to imply that the 
ACA is supporting criminal prosecution for copyright infringement.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] FYI: Campaign of Mis-Information

Jeff Waugh wrote:
The lack of context on welcomed the move does not change the entire nature
of the article, or significantly impact on its message.
 

I think the whole message of the first article (which is very different 
to the second article) is that all Australians are criminals when it 
comes to copyright law so don't complain when criminal copyright laws 
are introduced.  Nevermind the fact that copying-stuff-off-tv has never 
been a criminal act. 

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Re: Re: Copyright assignment + the GPL [Was: Streaming media servers]

2005-02-10 Thread QuantumG

Glen Turner wrote:
The only thing that is clear is that the developer needs a
high level of trust in the entity requesting the copyright
assignment.   After SCO I'm not sure that any of Ray Noorda's
current or previous activities engenders the necessary level
of trust for me.

Exactly.  I just happen not to trust anyone who declares they are going 
to accept my copyright and relicense it under a proprietary license.  If 
you wanna do that, fine.  On the other hand, if an open source project 
request that you assign your copyright to some entity and guarentees 
that they will not dual license your code under a proprietary license 
you should at least consider it.  It would be a lot better if they 
guarenteed (as the FSF does) that they will only use their ownership of 
the copyright to forward Free Software.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Streaming media servers

Shaun Butler wrote:
You could also try http://helixcommunity.org/. The open source version 
of helix Server will play REAL streaming protocols, but if you wanted 
to stream quicktime and Windows Media, you have to buy the commercial 
version :(.

I find it so scary that these companies which dual license have the 
boldness to require contributors to assign their copyright to them.  
I've only ever assigned my copyright to two groups in the past:  my 
employers, obviously, and the FSF.  I've never been worried about 
assigning my copyright to the FSF because I know they will ensure the 
software remains free (and with their gaggle of lawyers they can do a 
better job of that than I ever could).  But assigning your copyright to 
RealNetworks for the specific purpose so that they can dual license it, 
is just unpaid labor.  When you contribute to Helix you're not 
contributing to a community project, you're contributing to 
RealNetworks' bottom line.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Streaming media servers

Jeff Waugh wrote:
If the project has a sane license, then you are contributing to the commons
in perpetuity (almost, let's not get too finicky), so regardless of the
fortunes of the company (who are most likely doing the majority share of the
development work - why shouldn't they have a good bottom line?), the code
will be Free.
 

I think this completely boils down to a BSD vs GPL question.  The GPL 
and other copyleft licenses (like the RPSL claims to be) serve a purpose 
that the BSD does not and assigning your copyright to someone who then 
goes and uses it in a proprietary product weakens that.

Not that I have anything against the BSD.  The project I have spent the 
most time on is under a BSD like license.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

[SLUG] Re: Copyright assignment + the GPL [Was: Streaming media servers]

Jeff Waugh wrote:
I understand why this perspective has currency, but I disagree with it quite
strongly. The code is available under the GPL, the community has full GPL
rights, and you have a single copyright owner who is prepared to defend the
software and aggressively develop/improve the software because they have an
important investment in it.
 

Well the point of the GPL (over a BSD-like license) is to ensure that 
the source code for all distributed changes are available and are freely 
distributable.  So yes, although your specific changes are available to 
the community, all the changes that the new copyright owner distributes 
under a proprietary license are not.  You might be ok with this, but I 
think it's just paying lip service to the GPL without actually accepting 
the concept of software freedom.

As you put forward a number of bad situations that can arise from this, 
allow me to set up my nightmare scenario for you.

   1. You contribute a significant amount of work to Helix  (like a new 
feature).
   2. RealNetworks demands a copyright assignment.
   3. They put it into both the open source and proprietary distributions.
   4. They sell their proprietary distribution to customers.
   5. A customer finds a bug, wants a feature extension, etc.
   6. RealNetworks adds fixes the bug, or makes the feature extension 
in the proprietary distribution only.

That bug or feature extension is not free.  It is a proprietary 
extension of copylefted code.  That's crazy.  It's this kind of nonsense 
that specifically inspired the creation of copyleft licenses, but the 
license does not apply to the copyright holder, so the community 
suffers.  Now maybe you have no respect for copyleft licenses (it's ok 
to say so, I know a lot of BSD developers who don't) but if you do  
surely you've got to recognise that this is a travesty.

To me, choosing the BSD over the GPL is deliberately saying proprietary 
forks of this software are ok by me.  That's cool if that's what you 
want to do (I work on a project that specifically does this).  But 
choosing the GPL over the BSD is deliberately saying that they are not 
ok, and I don't see why an exception for one proprietary fork (that made 
by the copyright holder) is ok.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Re: Copyright assignment + the GPL [Was: Streaming media servers]

Jeff Waugh wrote:
Which is the case, even if the copyright owner distributes the software
under another license.
 

If they distribute exactly the same code under another license.  They 
hardly ever do this because the user has no interest in getting 
something under a proprietary license if they can get the same thing 
under a liberal license.  What they do is distribute the same code with 
proprietary extensions.  The user has the choice, use the code available 
under a liberal license and do without the features, or give up their 
freedom for the features that are only available under the proprietary 
license.  This is exactly what RealNetworks is doing with Helix. 

No, complete lack of software freedom inspired the creation of copyleft
licenses, not sensible business models built upon the foundations of
software freedom.
 

The BSD licenses were around long before the GPL was drafted.  RMS 
rejected the BSD because it failed to prevent this kind of proprietary 
forking.  At least that's my understanding of it.

The community benefits *more* from the availability of the software, source
and commitment of the organisation that owns its copyright than it would if
the source were closed.
 

But that's not the argument.  The argument is whether or not the 
community would benefit more from having someone like RealNetworks not 
demand copyright assignment.  I believe they would because any 
extensions RealNetworks makes to the work of contributors will remain free.

I'm sure the situation with RealNetworks is a lot more complicated than 
it appears.  For example, I'm sure their proprietary product includes 
some licensed technology that they obviously have no permission to 
integrate into the open source distribution.  On the other hand, they 
list things such as Ad-Serving Support and Live File Archiving as a 
features that is only in the proprietary version.  I have no idea what 
they entail, but I can imagine that people would want this and be 
willing to fore-go their freedom to get it. 

Suppose I wasn't happy with RealNetwork's ad-serving.. maybe my 
customers were complaining that the ads take up too much bandwidth or 
something.  I can't fix that.  Well boo-hoo, I gave up my freedom to get 
that silly feature so it's my own problem.  Say I decide to do something 
about it and write an ad-serving extension for the open source 
distribution.  Are RealNetworks going to accept my patch?  Presumably 
not.  That's a terrible situation for the community to be in.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Re: Re: Copyright assignment + the GPL [Was: Streaming media servers]

Matthew Palmer wrote:
Helix.  Which one do you want?  No, you can't answer option 1 because Real
isn't going to allow that.
 

Sorry, I think the correct answer is to fork Helix, implement all the 
features that people want.  That way when people choose to use Helix 
because they want to stream some prorietary format that cannot be put in 
the community distribution they will have to make the decision to do 
without the work of the community, encouraging them not to use 
proprietary formats in the first place.

I think the real question is, what does the community gain by giving 
their copyright to RealNetworks?  A proprietary distribution?  That's 
hardly a plus.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Re: Re: Copyright assignment + the GPL [Was: Streaming media servers]

Jeff Waugh wrote:
You're making a general statement (that copyright assignment is bad), but
 

Not at all.  My general statement is that assigning copyright to people 
who want that copyright assignment so they can make a proprietary 
distribution of the software is bad.  Assigning copyright to the FSF is 
the best thing in the world.  I would suggest that everyone do it.

Trent
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Re: Re: Copyright assignment + the GPL [Was: Streaming media servers]

Jeff Waugh wrote:
You're making a general statement (that copyright assignment is bad), but
hanging on to a specific example [1]. Look at other projects that require
copyright assignment, such as Evolution, and see whether the points you make
apply equally.
 


From the Evolution FAQ:
  http://forge.novell.com/modules/xoopsfaq/?cat_id=30%23q60
  What do I get for my copyright assignment?
  In exchange for your grant of copyright, Novell does two things. 
  First, we promise to use the GPL or another Debian-endorsed Free 
  license for your code (although we may also grant other licenses on 
  other terms).

  Second, we grant back to you as author all your original rights to 
  do what you want with your code, without any fees or
  restrictions  from us. This includes your right to relicense under 
  whatever terms you may desire. That is, your grant lets Novell use
  your code, but you keep your original ownership rights as well.

So in exchange for your copyright you get a license you wouldn't need if 
you didn't give them your copyright and you get the honour of Novel 
being able to relicense your code as they see fit (i.e., you get to 
watch them take away people's freedom using your work).  What a deal!

The only reason they can do it is because they control the repository.  
If GNU Arch was more widely used people wouldn't mind getting lots of 
patches from lots of different places and this kind of centralization 
wouldn't be something these companies could capitalize on.

Trent

--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Re: [SLUG] Re: Re: Copyright assignment + the GPL [Was: Streaming media servers]