FP 5.6 Windows Install

[Fairlight]

In the relative spacial/temporal region of
Mon, Sep 25, 2006 at 07:54:35AM -0400, Ed Olmstead achieved the spontaneous
generation of the following:
> It is the right one (setp.exe) - really weird behavior.  The good 
> money is they will tell me to download again in case there is some 
> file corruption from the download.

Which is a BS answer.  If there was corruption during the download, any
self-respecting archiving software will self-detect said corruption.  Most
setup utilities won't even unpack themselves if they're corrupt.

There's also the point that, unless you're using stop/resume type
downloaders that are so buggy that they mismatch their file offsets on
a resume, it's -almost- impossible to get a corrupted download.  TCP/IP
-guarantees- that the packets arrive, that they arrive (at the application
level) in the order in which they were sent, and that they're in-tact
(barring any kind of malicious injection).  If the downloads were over
FTP or HTTP, I'd put your realistic odds of a "corrupted download" at
about 0.001% or lower, given proper use of the software (ie., not using
ASCII mode when you should be using BINARY mode in FTP--and I -always- use
binary).  These are not the old xmodem days--and even then, there were
CRC's that prevented this kind of thing.

The sad state of the industry is that people will grasp at straws to
blame problems on anything except what is likely.  It doesn't matter how
illogical the argument is, for the straw-grabber it's something that puts
off the inevitable act of actually having to work to find the real problem.

In eleven years of commercially publishing my own software, I have yet to
-ever- tell someone they may have actually gotten a bad download, please
try again.  Actually, I'm more likely to have the enduser believe this
myth and try repeatedly to re-download things, so indoctrinated are people
by the industry to falsely believe this is a common occurance.  Nay, I
find any problem (assuming it's not user error from description) and I fix
it...period, the end.

mark->