2 Nov
2004
2 Nov
'04
11:36 a.m.
On Tue, Nov 02, 2004 at 06:41:46AM -0800, Mark Knecht wrote:
Good or bad the kernel
developers, be it Linus, or Alan, or who ever is running things at the
time, are free to make a change that effects many people. Those
changes are not voted on in any formal sense. They may meet the needs
to the overall group or they may not.
In the IEEE world there is no
'implementation'. Only a standard
document. Companies are left to develop the implementation themselves
and then, in the process, attempt to be compliant with the standards
and compatible with other devices that advertise their compliance with
the same standard. It's a very different approach.
I'm sure the contributors to IEEE standards committees all have their own
agendas and quite specific implementations in mind :-)
Sorry Mark, the rest of this isn't particularly a reply to your message,
just on the thread generally...
I'm glad we've got away from the "is it an OS or a kernel" discussion -
given the start of the thread was about presenting Linux-based OSS as a
platform to the rest of the world, that sort of nit-picking won't
impress anyone who just wants to do audio and knows what Windows is!
As for standards, back in the days when people were still asking
questions like "how compatible is Linux with standard UNIX?" a good
point that was made about standards: given a choice of different ways of
implementing something, the Linux (or any other OSS) designers would
always choose an open and widely used specification in favour of
anything closed and proprietary. This made a convincing case that Linux
was *more* standard than any particular proprietary system that called
itself UNIX.
It is well known, for example, that open source web browsers follow W3C
standards closely and their designers have no motivation to lock users into
proprietary "extensions". Or that Linux can read Windows file systems
but the converse has never been true. In every place where it matters,
GNU/Linux/OSS can and does fit in with existing standards, even
sometimes by reverse-engineering when those "standards" aren't published.
To suggest an open source program might change to something that breaks
your system overnight is bad publicity that fits right in with common
misconceptions about OSS. And it just isn't true: in the real world,
proprietary software does that much more. (MS Office file formats?)
--
Anahata
anahata(a)treewind.co.uk Tel: 01638 720444
http://www.treewind.co.uk Mob: 07976 263827