System N history correction (unbundling of System V)

Geoff Collyer geoff at utcs.UUCP
Sat Dec 22 08:38:22 AEST 1984


Michael Baldwin wrote:

> > >Last I heard, IWB is (a) separate, (b) extra cost, and (c) binary-only.

> And what's the problem with separate and extra
> cost?  That's the way you sell operating systems; everybody does that.
> It doesn't make sense to sell one monolithic system.  In SVR2, even the
> text processing tools are separate.  So what?

Doing something because everybody else does it is pretty mindless,
but that's what I've come to expect from
USG/USDL/what-ever-they-call-themselves-this-week.
That's the way *who* sells operating systems?

It's a sad day when UNIX is perceived as a monolithic system.
For you newcomers, UNIX was conceived as a reaction *against* monolithic
systems; it is a tribute to USG/USDL/...'s destructive power that they
have made System V a monolithic system or at least made some people see
it that way.

As System V is unbundled, I haven't noticed the price of the base system
dropping, only the creation of expensive ``work-benches'' which are simply
old code that's been hacked over by USG/USDL/...

I gather that the trend to unbundling will continue and we can look forward
to a base System V Release N consisting of a kernel and a collection of work-
benches, including the Concatenator's Work Bench (cat with options),
the Editor's Work Bench (ed), the Echoer's Work Bench (echo), the Logician's
Work Bench (true and false), the User's Work Bench (login), the Manual
Reader's Work Bench (man) and the Sleepy-head's Work Bench (sleep).



More information about the Comp.unix.wizards mailing list