This message is from the T13 list server.
Three points of note...
Personalities
-------------
Its important to keep a few critics around. It helps keep everyone else
honest. Even if you disagree with the issues raised, alternative
viewpoints should be given airtime.
Richard Stallman, the father of the GPL and free software, will yammer
on at length about how "all hardware/software should be free", right
down to the firmware and silicon RTL. The best response is to smile
politely, let him burn off steam ranting, and then continue on with your
day. People like Stallman help define the debate, even if you don't
agree with them.
Conspiracies
------------
The phrase "SATA secret society" is a blatant troll (i.e. fishing for
responses). Nonetheless its a valid if inflammatory criticism. SATA
development should be a lot more open.
However, one must be realistic: in this age of intellectual property
and competition, can it really be any more open than it already is? My
opinion is yes, but I'm likely in the minority.
Data corruption
---------------
I think that SATA is better than PATA at detecting corruption. So, I am
skeptical of reports of data corruption in the base technology.
More likely is data corruption and/or missed errors due to legacy OS
driver implementations which are not fully SATA-aware. The various
early SATA host controllers could be described as being "98% compatible"
with PATA, rather than 100%.
Also, while (as stated above) I feel SATA detects corruption better than
PATA, my gut feeling is that SATA generates far more cable errors. If
the OS driver error handling code is correct, no problem. If your error
handling code paths are rarely travelled, SATA could reveal bugs in your
code.
Jeff
