[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: LCC and blobs

Brian Thomas Sniffen <bts@alum.mit.edu> writes:

> Anthony DeRobertis <anthony@derobert.net> writes:
>> Brian Thomas Sniffen wrote:
>>> That's not software.  That's firmware, at best -- you can look at it
>>> as software, but then you don't get to distribute any drivers.  It is
>>> also internally consistent to think of chips as hardware and
>>> distribute drivers appropriately.  It is never consistent to think of
>>> files on disk as anything but software.
>> Hmmm, I have a CF card. Upon it are files, and in every meaningful way
>> it is a disk. Therefor, that data is software.
>> Yet, CF is actually chips --- often the same chips as used to hold
>> firmware distributed with hardware. Thus, it's all hardware.
> Sure.  It's on a medium for software exchange, thus it's software.  If
> it were an integral component of a device, it'd be hardware.

If it is glued to the socket, does it become hardware?

> I've never been confused when looking at such things; the closest
> case I've found to confusing is the MP3 player, which has its OS on
> disk.  I'm inclined to say that it's software to the MP3 player, an
> architecture which Debian does not support.  It's hardware, a drive,
> to the (Intel?) Debian-supported PC to which it's connected.

Why is this different from the SCSI controller board, whose CPU (and
related components) is also an architecture not supported by Debian?
Does it matter whether it connects to the PC by the PCI bus directly,
or over USB?

Måns Rullgård

Reply to: