On 03 Dec at 15:34, Charlie Decker of 1:323/1212 wrote to Steve Steffler:
CD> Steve Steffler
CD> blew off some steam in  about Conference Rules ....:
DC>> It is possible to damage hardware through software, it'd (usually)
DC>> just take a really stupid user for it to actually work :-)
PW>> Perhaps you can tell me what FORMAT.EXE does, David.
SS>> No such program exists.  I have a FORMAT.COM, but no .EXE...
SS>> And by the way, the FORMAT command doesn't damage hardware.
CD> Have a sudden power loss while using it and get back to us... :>
At *worst*, it may cause a corruption of the sector header on the drives 
platter. This is remedied by Low Level Formating the drive. 
The newer drives do not allow the servo platter to be written to so even this 
is not possible these days.
Otherwise, corruption to the data written to the drive is possible. It is not 
physical damage so cannot be construed as hardware damage. 
Re-formatting that track and re-writing the data will fix the problem proving 
once again that the damage was not to the hardware.
Also, FORMAT does not write to a hard disk  until it has checked the 
integrity of the drive where it then writes a brand new boot sector, root 
directory and FAT(s). 
That is why a formatted hard drive can be unformatted. Without using special 
utilities such as Norton, it is possible to recover everything except the 
root directory after an accidental format.
The definition of hardware damage must be actual physical damage to the 
hardware.
No one has =ever= shown anyone that such monsters exist that will cause this.
After many years, we are all still waiting for a working example.
Here is an open challenge.
==========================
If anyone sends me an executable that is able to successfully physically 
damage hardware on my machine, I will announce this to the whole FTSC and 
Internet community.
The rules are simple. 
The file is =sent= to me. Without any changes to my hardware which I use 
routinely every day, I will run the proposed rogue program.
The machine is a generic IBM clone machine. 
Actual machine details should not be necessary because any program that is 
able to cause physical damage should do so to any machine without favour and 
without regard to what actual brands/models are in use.
I have three machines here that use different hardware entirely. I will run 
the submitted program on all three.
The program will be allowed ten seconds to complete it's task. That should be 
sufficient to prove that the damage is a result of the program and not a 
random event caused by wear and tear.
I am qualified to diagnose any damage caused to the machine and it is in 
fact, my job as both a repairer and a consultant that determines that I am 
able to do so. 
I have component level experience including design and manufacture of 
electronic devices over the past 30+ years. 
I have extensive acadaemic qualifications in the electronics industry.
Now let this go until someone is able to provide me with code that will do 
this (so far) mythical physical damage.
I won't hold my breath waiting for a working example.
The last time this challenge was thrown open there was no reponse.
Regards,
Chris Maddock
chrism@bbs.st.net.au
3:640/302@fidonet
--- Msged/386 4.20 beta 2
---------------