[UFO Chicago] debian woody on an imac rev. b -- success! (fwd)

[presence]

On Fri, 10 Jan 2003, Larry Garfield wrote:

[cut]

> Modify: Bradley M. Kuhn, an individual of my aquaintance, was once
> working in a computer lab back in the early 1990s.  All the systems were
> running GNU/Linux, an old 1.x version.  They wanted to keep users from
> Ctrl+Alt+Deleting and restarting the system.  (Security reasons.)  There
> was no BIOS or boot flag for it.  However, they had access to the source
> code.  So Brad went into the source code, MODIFIED it to comment out the
> block of code that responded to the Ctrl+Alt+Del command, recompiled the
> kernel, and installed it on all of the lab machines.  That effectively
> and completely prevented users from Ctrl+Alt+Deleting to reboot the
> system.  Had he not had access to the source code or not had permission
> to MODIFY it, he would have not been able to complete the task.

Hmm. If it had been written right to start with, maybe the assumption that
control+alt+delete should reboot the machine would never have been made.

How many people do you think have written the same patch for that, over
and over and over again?

> Redistribute: In most cases, this takes two forms.
>
> 1) Take a snippit of the code in the program that does one task and drop
> it into another program to do the same task in a different program.
> (Eg, a sorting algorithm, a thread parser, etc.)

No amazing code was made to make the imac mentioned play mp3s better.

> 2) Immediate fix.  A bug in man, if released, can be fixed at any time
> by any of a few hundred thousand people and uploaded.  Such fixes are

Hmm, maybe with quality control, it would never have been created to start
with. Let's all just go out and do our own thing and then we cann all
write our own patchs later on. This burns no time at all or anything.

> usually in place in a matter of hours, days at the most.  If the code is
> not available to be modified and REDISTRIBUTED (in this case, back to
> the original author), then there are perhaps 5 people who can fix it and
> upload it, and they generally have other things on their minds.
> Everyone then gets a very quick turnaround time on bug fixes.

Fast hasty patches made on a whim never introduce new problems.

> What you are talking about is "forking", which does happen but not as
> often as you would expect.  (Surprisingly little, actually.)

Yup, it's never happened.

> I'm not even a Free Software advocate, and even I understand these
> things. :-)
>
> --
> Larry Garfield			AIM: LOLG42
> larry@garfieldtech.com		ICQ: 6817012
>
> "The world's most dangerous terrorist is at large in the US.  He has his
> sights set on making the entire American population cower in fear.  He
> has ordered his people to assassinate American citizens.  Do you know
> him?  His name is George W. Bush."

Wait, I fear waiting for a solid patch from the people who write the
software in the first place. Let me head to the nearest com-bloc FTP
server for version X of somebody's version of a quik fix.

KEN
"Linux God"