[geeks] q: good source for leather book bag?

[Mike Meredith]

On Wed, 3 Jun 2009 16:52:55 -0400 (EDT), nate at portents.com wrote:
> And you've illustrated one of the big problems with
> UNIX/Unix/Linux/BSD/distributions - it's kind of a mess.  I think the
> chart says it all:
> 
> http://www.levenez.com/unix/

Well not that I disagree that it's a mess, but the chart does cover
nearly 40 years of IT history and attempts to cover _everything_
(and doesn't) including stuff that really isn't Unix at all (such as
Mach which is only included because most Mach installations had a Unix
single server usually based around BSD, and is effectively the origin
of OS X).

If you look at what's available in the mainstream today, there really
isn't that much confusion at all. 

> You've shown how hard it is to have a conversation about it without
> splitting some very fine hairs - differing letter cases on identically
> lettered words implying different meanings one of which has no formal
> definition but rather an implied operational definition that only
> people who work in that field would ever pick up on.

The whole case issue is merely illustrating the stupidity of the
trademark issue where UNIX(tm) _became_ merely a badge of compliance
with a set of API standards that had nothing to do with the operating
system itself.

To me, "Unix" is an operating system directly descended from the
original Bell Labs version ... which does include the BSD family. BSD
code eventually replaced the original Bell Labs code, but probably not
at a very different rate to the rate at which AT&T replaced the Bell
Labs code. Originally of course this would have been UNIX(tm). I'd also
(believe it or not) include variants such as Xenix that weren't allowed
to call themselves UNIX(tm) because although their originators licensed
the UNIX(tm) source code, they didn't license the right to use the
UNIX(tm) "label" (I _think_ that's right).

Just as important are the many operating systems that tried to be
reasonably close to Unix, but didn't qualify ... because they
were written without access to the UNIX source code. There's many
people who wanted access to Unix but for one reason or another couldn't
justify the high cost of a "real" Unix or needed features that took a
while to get into "real" Unix (such as real time scheduling).

> Curiously, I think this un-coordinated overly bifurcated giant glob of
> software lubering through time indirectly supports Microsoft.  Until
> all this "Unix" stuff unifies, streamlines and simplifies the
> families, we will have Microsoft be dominant.

It would also help if the suits realised that just because any idiot
can fly Windows, doesn't mean your servers should be run by idiots. The
easiest way to avoid idiots is to get a Unix administrator*

*: That isn't to say that all Windows server administrators are idiots
   or that there aren't any Unix administrators who aren't idiots, but
   Windows does seem to have a greater attraction to idiots ("it's
   easy" after all :) ).


-- 
Mike Meredith (http://zonky.org/)
 By the way, you DON'T want to see what a meat layer buffer overrun
 looks like.... (mjr on fw-wiz)