Xref: utzoo comp.unix.questions:12494 comp.society.futures:1120
Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!csd4.milw.wisc.edu!lll-winken!uunet!auspex!guy
From: guy@auspex.UUCP (Guy Harris)
Newsgroups: comp.unix.questions,comp.society.futures
Subject: Re: Future at Berzerkeley
Keywords: Leading Edge == Bottom Line
Message-ID: <1300@auspex.UUCP>
Date: 27 Mar 89 00:15:52 GMT
References: <4572@pt.cs.cmu.edu>
Reply-To: guy@auspex.UUCP (Guy Harris)
Organization: Auspex Systems, Santa Clara
Lines: 84

>BSD will have more innovative features than SysV for
>a long time,

I presume then that some future BSD release will have a dynamic linking
mechanism, complete with a programmatic interface to the run-time loader,
so that you can map a dynamically-linked object into your address space,
look up functions in that object by name and get pointers to them, and
unmap the object from your address space?  I expect S5R4 to have
that....

(It'll also be interesting to see whether 4.4BSD, with an "mmap" that
lets you map files into your address space, comes out before S5R4, with
an "mmap" that lets you map files into your address space.)

Now, CMU already has something of that general nature that runs on BSD
systems, as I understand it (Andrew uses it, as I remember), so in that
sense BSD will have it, even though it may not be on the tape you get
from Berkeley. 

>Witness AFS, the Andrew File System: the first nationwide
>file system.  It runs under BSD, and I don't think it'll be
>on SysV machines for a while.  But, I could be wrong.

It depends; if a version of AFS is made whose kernel-level functions
plug into a VFS-type interface, it will be possible to plug it into an
S5R4 system.  Does AFS run under an unmodified BSD?  If not, then the
real question to ask is "how easy is it to modify BSD to support it vs.
how easy is it to modify S5 to support it".

>BSD code is more accessable; if AT&T wants major innovations
>under SysV, they are going to have to be more easy-going 
>with the sources and software hooks.

I'm not sure what you mean by "more easy-going... with the software
hooks"; if you have the source, you can put in whatever hooks you want,
and I suspect the number of hooks of that flavor that come with the
system off the distribution tape won't be higher for BSD than S5.

I think the bottom line for BSD vs. S5 is:

	1) availability of source.   Unless S5 source is missing critical
	   pieces that are present in BSD releases (which could
	   conceivably happen), then unless Berkeley comes out with a
	   completely AT&T-code-free release (which they may well do, at
	   some point), you'll have to buy S5 source anyway to get BSD
	   source (unless you already have a licence sufficent to get
	   you BSD releases "in perpetuity", which I suspect many,
	   possibly most, research institutions already have).

	   Given the parenthetical phrases listed there, it could well
	   be that BSD source is more available to research institutions
	   than S5 source is.

	2) familiarity.  I have the impression that BSD dominates in
	   research institutions, especially universities; it may be
	   that adapting either projects or people to an S5 release may
	   require sufficient effort so as to discourage such
	   adaptation.  The barrier may reduce with S5R4.

	   Of course, some of the machines in those institutions may be
	   running the vendor's OS, rather than BSD; even for vendor's
	   OSes that come, in part, from BSD, there are probably
	   barriers of that sort.  (SunOS is *not* BSD, for example. 
	   It's not *intended* to be "a BSD port to Suns".)

The issue of "what research projects, or innovative features, run under
BSD vs. System V" - this basically refers to features that *don't* come
on the distribution tape, e.g. the Andrew File System - is probably
governed by the issues above. 

Of course, this also brings up the issue of "will any of those features
get *on* the distribution tape."  It's conceivable that they'll get on
the BSD tape more easily than they'll get on the S5 tape.

My personal guess, for what it's worth (which isn't a hell of a lot;
from what I've seen, I don't put much stock in predictions of the future
in this field), is that some research flavor of UNIX will continue to be
popular in universities and other research institutions.  I don't know
whether it'll be BSD, or Mach, or something else; I suspect it'll be
less stable than S5-from-AT&T, as suggested by others, and that its
relative lack of stability will be precisely one of those features that
*makes* it popular - for instance, it'll be easier to get innovations
onto the distribution tape.  It may be easier to get source for it,
which may help as well.