/sys/doc/ Documentation archive

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

Re: Apple

thus, so far, the possibilities for future macos:

   (0) rewrite in limbo
   (1) rewrite in java
   (2) emulate under beos
   (3) emulate under nt
   (4) emulate under mae
	(now on solaris + hpux)

notes: (0) is missing a powermac port;
	   prototypical mass market limbo host
	   is a "sealed" net pc running inferno.
       (1) now available as a layer on a realtime
	   kernel (os-9); mass market prototypes
	   use javachips and kona.
       (2)/(3) have the advantage that ports
	   exist for underlying hardware which
	   apple currently sells.
       (4) porting mae makes for other milieus
           (sgi irix, ibm aix, etc.); rehosting
	   solaris from to powermac is still
	   possible (but existing powerpc
           port is other-endian); mae rewritten
	   in limbo, with X replaced by tcl
	   could do much towards #0 

apple merger candidates remain those with:

	(0) money ($5-10 billion)
	(1) existing weak consumer brand
	(2) urge to distribute new software
	    to millions.

notes: of the announced attempts (ibm, sun, oracle, at&t),
sun & oracle have the weak brand, but sun wants 
telcos to oem java [source -- mcnealy, shareholder meeting].
of unannounced possibilities, sgi has no money, microsoft has
antitrust problems as well as lacking #1, some faceless rich
foreign combine may have all three, sony & disney are disqualified
on brand.  lucent *does* make sense, but my ex-boss
and former head of apple r&d (david nagel) went to at&t,
not lucent.

if lucent is rebuffed, a natural choice for a
consumer-branded distributor for limbo/inferno/styx
machines must surely be

	virgin

i.e. richard branson's empire (airline, cola, vodka, internet service).

tagline: inside every virgin boils an inferno

--jaw