[Wolves] Remember that HPT372 RAID thing I was on about?

[Aquarius]

Dan spoo'd forth:
> Luckily the device numbers *are* 
> contained within a readable header in the download so I could simply 
> change the device number of the 372A to that of the 374 - now it's 
> recognised as two 372As, and actually works suprisingly well. 

*cough* hack! *cough*

> Needless to say, Highpoint are not high on my list of favourite 
> companies at the moment.  Still, at least they do support Linux to an 
> extent I suppose.

I'm unclear whether making bitchy noises at or about them will (a) let
them know that what they're providing isn't up to the challenge (good)
or (b) encourage them to just drop support as too complex (bad). RMS
would definitely take view a, in that if you're doing bad support and
hearing complaints makes you drop it altogether then maybe the free
software community doesn't need you anyway. On the other hand, someone
who considers themselves a realist would go for (b) and not say
anything. I'm inclined to suggest that a minor mail to them explaining
your problems, without making any accusations, might help, because they
could well be genuinely unaware of this problem and might leap to fix
it. Actually fixing the problem and sending them the patch would be
better, but you can't do that if it's binary -- if you're feeling
particularly narked about it then tell the Open Source Foundation about
how Highpoint are using the words in vain and see if that helps...?

Aq.

-- 
Why am I not surprised that Microsoft would make these things HTTP
Equivs? What next? Extending HTTP 1.1 so that "GET" is followed by "or
else"?
           -- Steven Champeon <schampeo@hesketh.com>, webdesign-l