<s>
The	O
Tanenbaum	B-Language
–	I-Language
Torvalds	I-Language
debate	I-Language
was	O
a	O
written	O
debate	O
between	O
Andrew	O
S	O
.	O
Tanenbaum	O
and	O
Linus	O
Torvalds	O
,	O
regarding	O
the	O
Linux	B-Operating_System
kernel	I-Operating_System
and	O
kernel	B-Operating_System
architecture	B-General_Concept
in	O
general	O
.	O
</s>
<s>
Tanenbaum	O
,	O
the	O
creator	O
of	O
Minix	B-Operating_System
,	O
began	O
the	O
debate	O
in	O
1992	O
on	O
the	O
Usenet	B-Application
discussion	O
group	O
,	O
arguing	O
that	O
microkernels	B-Operating_System
are	O
superior	O
to	O
monolithic	B-Operating_System
kernels	I-Operating_System
and	O
therefore	O
Linux	B-Operating_System
was	O
,	O
even	O
in	O
1992	O
,	O
obsolete	O
.	O
</s>
<s>
While	O
the	O
debate	O
initially	O
started	O
out	O
as	O
relatively	O
moderate	O
,	O
with	O
both	O
parties	O
involved	O
making	O
only	O
banal	O
statements	O
about	O
kernel	B-Operating_System
design	I-Operating_System
,	O
it	O
grew	O
progressively	O
more	O
detailed	O
and	O
sophisticated	O
with	O
every	O
round	O
of	O
posts	O
.	O
</s>
<s>
Besides	O
just	O
kernel	B-Operating_System
design	I-Operating_System
,	O
the	O
debate	O
branched	O
into	O
several	O
other	O
topics	O
,	O
such	O
as	O
which	O
microprocessor	B-Architecture
architecture	B-General_Concept
would	O
win	O
out	O
over	O
others	O
in	O
the	O
future	O
.	O
</s>
<s>
Besides	O
Tanenbaum	O
and	O
Torvalds	O
,	O
several	O
other	O
people	O
joined	O
the	O
debate	O
,	O
including	O
Peter	O
MacDonald	O
,	O
an	O
early	O
Linux	B-Operating_System
kernel	I-Operating_System
developer	O
and	O
creator	O
of	O
one	O
of	O
the	O
first	O
distributions	B-Application
,	O
Softlanding	B-Operating_System
Linux	I-Operating_System
System	I-Operating_System
;	O
David	O
S	O
.	O
Miller	O
,	O
one	O
of	O
the	O
core	O
developers	O
of	O
the	O
Linux	B-Operating_System
kernel	I-Operating_System
;	O
and	O
Theodore	O
Ts'o	O
,	O
the	O
first	O
North	O
American	O
Linux	B-Operating_System
kernel	I-Operating_System
developer	O
.	O
</s>
<s>
The	O
debate	O
opened	O
on	O
January	O
29	O
,	O
1992	O
,	O
when	O
Tanenbaum	O
first	O
posted	O
his	O
criticism	O
on	O
the	O
Linux	B-Operating_System
kernel	I-Operating_System
to	O
,	O
noting	O
how	O
the	O
monolithic	B-Operating_System
design	O
was	O
detrimental	O
to	O
its	O
abilities	O
,	O
in	O
a	O
post	O
titled	O
"	O
LINUX	B-Language
is	I-Language
obsolete	I-Language
"	O
.	O
</s>
<s>
While	O
he	O
initially	O
did	O
not	O
go	O
into	O
great	O
technical	O
detail	O
to	O
explain	O
why	O
he	O
felt	O
that	O
the	O
microkernel	B-Operating_System
design	O
was	O
better	O
,	O
he	O
did	O
suggest	O
that	O
it	O
was	O
mostly	O
related	O
to	O
portability	B-Architecture
,	O
arguing	O
that	O
the	O
Linux	B-Operating_System
kernel	I-Operating_System
was	O
too	O
closely	O
tied	O
to	O
the	O
x86	B-Operating_System
line	O
of	O
processors	O
to	O
be	O
of	O
any	O
use	O
in	O
the	O
future	O
,	O
as	O
this	O
architecture	B-General_Concept
would	O
be	O
superseded	O
by	O
then	O
.	O
</s>
<s>
To	O
put	O
things	O
into	O
perspective	O
,	O
he	O
mentioned	O
how	O
writing	O
a	O
monolithic	B-Operating_System
kernel	I-Operating_System
in	O
1991	O
is	O
"	O
a	O
giant	O
step	O
back	O
into	O
the	O
1970s	O
"	O
.	O
</s>
<s>
He	O
did	O
so	O
a	O
day	O
later	O
,	O
arguing	O
that	O
MINIX	B-Operating_System
has	O
inherent	O
design	O
flaws	O
(	O
naming	O
the	O
lack	O
of	O
multithreading	B-General_Concept
as	O
a	O
specific	O
example	O
)	O
,	O
while	O
acknowledging	O
that	O
he	O
finds	O
the	O
microkernel	B-Operating_System
kernel	B-Operating_System
design	I-Operating_System
to	O
be	O
superior	O
"	O
from	O
a	O
theoretical	O
and	O
aesthetical	O
"	O
point	O
of	O
view	O
.	O
</s>
<s>
He	O
also	O
claimed	O
that	O
since	O
he	O
was	O
developing	O
the	O
Linux	B-Operating_System
kernel	I-Operating_System
in	O
his	O
spare	O
time	O
and	O
giving	O
it	O
away	O
for	O
free	O
(	O
Tanenbaum	O
's	O
MINIX	B-Operating_System
was	O
not	O
free	O
at	O
that	O
time	O
)	O
,	O
Tanenbaum	O
should	O
not	O
object	O
to	O
his	O
efforts	O
.	O
</s>
<s>
Furthermore	O
,	O
he	O
mentioned	O
how	O
he	O
developed	O
Linux	B-Operating_System
specifically	O
for	O
the	O
Intel	B-General_Concept
80386	I-General_Concept
because	O
it	O
was	O
partly	O
intended	O
as	O
a	O
learning	O
exercise	O
for	O
Torvalds	O
himself	O
;	O
while	O
he	O
conceded	O
that	O
this	O
made	O
the	O
kernel	B-Operating_System
itself	O
less	O
portable	O
than	O
MINIX	B-Operating_System
,	O
he	O
asserted	O
that	O
this	O
was	O
an	O
acceptable	O
design	O
principle	O
,	O
as	O
it	O
made	O
the	O
application	B-General_Concept
programming	I-General_Concept
interface	I-General_Concept
simpler	O
and	O
more	O
portable	O
.	O
</s>
<s>
For	O
this	O
reason	O
,	O
he	O
stated	O
,	O
"	O
linux	B-Operating_System
is	O
more	O
portable	O
than	O
minix.	O
"	O
</s>
<s>
Following	O
Linus	O
 '	O
reply	O
,	O
Tanenbaum	O
argued	O
that	O
the	O
limitations	O
of	O
MINIX	B-Operating_System
relate	O
to	O
him	O
being	O
a	O
professor	O
,	O
stating	O
the	O
requirement	O
for	O
the	O
system	O
to	O
be	O
able	O
to	O
run	O
on	O
the	O
rather	O
limited	O
hardware	O
of	O
the	O
average	O
student	O
,	O
which	O
he	O
noted	O
was	O
an	O
Intel	O
8088-based	O
computer	O
,	O
sometimes	O
even	O
without	O
a	O
hard	B-Device
drive	I-Device
.	O
</s>
<s>
Linux	B-Operating_System
was	O
,	O
at	O
that	O
time	O
,	O
specifically	O
built	O
for	O
the	O
Intel	B-General_Concept
386	I-General_Concept
,	O
a	O
significantly	O
more	O
powerful	O
(	O
and	O
expensive	O
)	O
processor	O
.	O
</s>
<s>
Tanenbaum	O
also	O
specifically	O
states	O
"	O
...	O
as	O
of	O
about	O
1	O
year	O
ago	O
,	O
there	O
were	O
two	O
versions	O
[	O
of	O
MINIX ]	O
,	O
one	O
for	O
the	O
PC	O
(	O
360K	O
diskettes	O
)	O
and	O
one	O
for	O
the	O
286/386	O
(	O
1.2M	O
)	O
.	O
</s>
<s>
The	O
PC	O
version	O
was	O
outselling	O
the	O
286/386	O
version	O
by	O
2	O
to	O
1.	O
"	O
</s>
<s>
He	O
noted	O
that	O
even	O
though	O
Linux	B-Operating_System
was	O
free	O
,	O
it	O
would	O
n't	O
be	O
a	O
viable	O
choice	O
for	O
his	O
students	O
,	O
as	O
they	O
would	O
not	O
be	O
able	O
to	O
afford	O
the	O
expensive	O
hardware	O
required	O
to	O
run	O
it	O
,	O
and	O
that	O
MINIX	B-Operating_System
could	O
be	O
used	O
on	O
"	O
a	O
regular	O
4.77	O
MHz	O
PC	O
with	O
no	O
hard	O
disk.	O
"	O
</s>
<s>
To	O
this	O
,	O
Kevin	O
Brown	O
,	O
another	O
user	O
of	O
the	O
Usenet	B-Application
group	O
,	O
replied	O
that	O
Tanenbaum	O
should	O
not	O
complain	O
about	O
Linux	B-Operating_System
's	O
ties	O
to	O
the	O
386	B-General_Concept
architecture	B-General_Concept
,	O
as	O
it	O
was	O
the	O
result	O
of	O
a	O
conscious	O
choice	O
rather	O
than	O
lack	O
of	O
knowledge	O
about	O
operating	O
system	O
design	O
,	O
stating	O
"	O
...	O
an	O
explicit	O
design	O
goal	O
of	O
Linux	B-Operating_System
was	O
to	O
take	O
advantage	O
of	O
the	O
special	O
features	O
of	O
the	O
386	B-General_Concept
architecture	B-General_Concept
.	O
</s>
<s>
He	O
also	O
stated	O
that	O
designing	O
a	O
system	O
specifically	O
for	O
cheap	O
hardware	O
would	O
cause	O
it	O
to	O
have	O
portability	B-Architecture
problems	O
in	O
the	O
future	O
.	O
</s>
<s>
Despite	O
the	O
fact	O
that	O
MINIX	B-Operating_System
did	O
not	O
fully	O
support	O
the	O
newer	O
hardware	O
,	O
Tanenbaum	O
argued	O
that	O
since	O
the	O
x86	B-Operating_System
architecture	I-Operating_System
would	O
be	O
outdone	O
by	O
other	O
architecture	B-General_Concept
designs	O
in	O
the	O
future	O
,	O
he	O
did	O
not	O
need	O
to	O
address	O
the	O
issue	O
,	O
noting	O
"	O
Of	O
course	O
5	O
years	O
from	O
now	O
that	O
will	O
be	O
different	O
,	O
but	O
5	O
years	O
from	O
now	O
everyone	O
will	O
be	O
running	O
free	O
GNU	B-Operating_System
on	O
their	O
200	O
MIPS	O
,	O
64M	O
SPARCstation-5.	O
"	O
</s>
<s>
He	O
stated	O
that	O
the	O
Linux	B-Operating_System
kernel	I-Operating_System
would	O
eventually	O
fall	O
out	O
of	O
style	O
as	O
hardware	O
progressed	O
,	O
due	O
to	O
it	O
being	O
so	O
closely	O
tied	O
to	O
the	O
386	B-General_Concept
architecture	B-General_Concept
.	O
</s>
<s>
The	O
386	B-General_Concept
processor	O
was	O
then	O
the	O
most	O
widespread	O
chip	O
"	O
by	O
several	O
times	O
"	O
,	O
according	O
to	O
participant	O
Kevin	O
Brown	O
,	O
with	O
the	O
486	O
used	O
in	O
high-end	O
computers	O
,	O
the	O
286	B-General_Concept
almost	O
obsolete	O
,	O
and	O
the	O
World	O
Wide	O
Web	O
not	O
yet	O
widely	O
used	O
.	O
</s>
<s>
One	O
of	O
Tanenbaum	O
's	O
arguments	O
against	O
Linux	B-Operating_System
was	O
that	O
it	O
was	O
too	O
closely	O
tied	O
to	O
the	O
x86	B-Operating_System
architecture	I-Operating_System
and	O
instruction	O
set	O
,	O
which	O
he	O
regarded	O
as	O
a	O
mistake	O
.	O
</s>
<s>
Modern	O
Linux	B-Operating_System
now	O
has	O
a	O
more	O
portable	O
codebase	O
and	O
has	O
been	O
ported	O
to	O
many	O
other	O
processor	O
architectures	O
.	O
</s>
<s>
Another	O
recurring	O
topic	O
in	O
the	O
debate	O
discusses	O
alternatives	O
to	O
Linux	B-Operating_System
and	O
MINIX	B-Operating_System
,	O
such	O
as	O
GNU	B-Operating_System
(	O
Hurd	B-Application
)	O
and	O
4.4BSD	B-Operating_System
.	O
</s>
<s>
Tanenbaum	O
suggested	O
the	O
former	O
in	O
his	O
first	O
post	O
,	O
stating	O
that	O
unlike	O
Linux	B-Operating_System
,	O
it	O
was	O
a	O
"	O
modern	O
"	O
system	O
"	O
.	O
</s>
<s>
In	O
his	O
second	O
post	O
,	O
he	O
mentioned	O
that	O
"	O
...	O
5	O
years	O
from	O
now	O
everyone	O
will	O
be	O
running	O
free	O
GNU	B-Operating_System
on	O
their	O
200	O
MIPS	O
,	O
64M	O
SPARCstation-5	B-Architecture
"	O
.	O
</s>
<s>
Several	O
debaters	O
disagreed	O
that	O
GNU	B-Operating_System
was	O
a	O
suitable	O
alternative	O
.	O
</s>
<s>
Kevin	O
Brown	O
called	O
it	O
vaporware	B-Application
,	O
and	O
stated	O
that	O
Linux	B-Operating_System
would	O
likely	O
benefit	O
from	O
the	O
x86	B-Operating_System
architecture	I-Operating_System
which	O
would	O
continue	O
to	O
be	O
common	O
and	O
become	O
more	O
accessible	O
to	O
a	O
general	O
audience	O
.	O
</s>
<s>
Theodore	O
Ts'o	O
,	O
an	O
early	O
Linux	B-Operating_System
contributor	O
,	O
said	O
that	O
while	O
a	O
microkernel	B-Operating_System
approach	O
would	O
have	O
benefits	O
,	O
"	O
...	O
Linux	B-Operating_System
is	O
here	O
,	O
and	O
GNU	B-Operating_System
isn't—and	O
people	O
have	O
been	O
working	O
on	O
Hurd	B-Application
for	O
a	O
lot	O
longer	O
than	O
Linus	O
has	O
been	O
working	O
on	O
Linux	B-Operating_System
"	O
.	O
</s>
<s>
Torvalds	O
,	O
aware	O
of	O
GNU	B-Operating_System
's	O
efforts	O
to	O
create	O
a	O
kernel	B-Operating_System
,	O
stated	O
"	O
If	O
the	O
GNU	B-Application
kernel	I-Application
had	O
been	O
ready	O
last	O
spring	O
,	O
I	O
'd	O
not	O
have	O
bothered	O
to	O
even	O
start	O
my	O
project	O
:	O
the	O
fact	O
is	O
that	O
it	O
was	O
n't	O
and	O
still	O
is	O
n't	O
.	O
"	O
</s>
<s>
4.4BSD-Lite	O
would	O
not	O
be	O
available	O
until	O
two	O
years	O
later	O
due	O
to	O
the	O
USL	B-Operating_System
v	I-Operating_System
.	I-Operating_System
BSDi	I-Operating_System
lawsuit	O
,	O
filed	O
by	O
T	O
's	O
subsidiary	O
Unix	B-Application
System	I-Application
Laboratories	O
against	O
Berkeley	O
Software	O
Design	O
,	O
which	O
pertained	O
to	O
the	O
intellectual	O
property	O
related	O
to	O
UNIX	B-Application
.	O
</s>
<s>
The	O
lawsuit	O
slowed	O
development	O
of	O
the	O
free-software	O
descendants	O
of	O
BSD	B-Operating_System
for	O
nearly	O
two	O
years	O
while	O
their	O
legal	O
status	O
was	O
in	O
question	O
.	O
</s>
<s>
As	O
Linux	B-Operating_System
did	O
not	O
have	O
such	O
legal	O
ambiguity	O
,	O
systems	O
based	O
on	O
it	O
gained	O
greater	O
support	O
.	O
</s>
<s>
A	O
settlement	O
between	O
USL	B-Operating_System
v	I-Operating_System
.	I-Operating_System
BSDi	I-Operating_System
was	O
reached	O
in	O
January	O
1994	O
,	O
and	O
4.4BSD	B-Operating_System
was	O
released	O
in	O
June	O
.	O
</s>
<s>
(	O
While	O
the	O
final	O
release	O
was	O
in	O
1995	O
,	O
several	O
free	O
versions	O
based	O
on	O
this	O
version	O
have	O
been	O
maintained	O
since	O
,	O
including	O
FreeBSD	B-Operating_System
,	O
DragonFly	B-Application
BSD	I-Application
,	O
OpenBSD	B-Operating_System
and	O
NetBSD	B-Device
.	O
)	O
</s>
<s>
The	O
book	O
claims	O
that	O
Linux	B-Operating_System
was	O
initially	O
illegally	O
copied	O
from	O
MINIX	B-Operating_System
.	O
</s>
<s>
MINIX	B-Operating_System
was	O
only	O
a	O
kind	O
of	O
fun	O
hobby	O
for	O
me	O
.	O
</s>
<s>
I	O
wrote	O
MINIX	B-Operating_System
because	O
I	O
wanted	O
my	O
students	O
to	O
have	O
hands-on	O
experience	O
playing	O
with	O
an	O
operating	O
system	O
.	O
</s>
<s>
After	O
AT&T	O
forbade	O
teaching	O
from	O
John	O
Lions	O
 '	O
book	O
,	O
I	O
decided	O
to	O
write	O
a	O
UNIX-like	O
system	O
for	O
my	O
students	O
to	O
play	O
with	O
.	O
</s>
<s>
I	O
was	O
not	O
trying	O
to	O
replace	O
GNU/HURD	B-Application
or	O
Berkeley	B-Operating_System
UNIX	I-Operating_System
.	O
</s>
<s>
I	O
just	O
wanted	O
to	O
show	O
my	O
students	O
and	O
other	O
students	O
how	O
you	O
could	O
write	O
a	O
UNIX-like	O
system	O
using	O
modern	O
technology	O
.	O
</s>
<s>
A	O
lot	O
of	O
other	O
people	O
wanted	O
a	O
free	O
production	O
UNIX	B-Application
with	O
lots	O
of	O
bells	O
and	O
whistles	O
and	O
wanted	O
to	O
convert	O
MINIX	B-Operating_System
into	O
that	O
.	O
</s>
<s>
I	O
was	O
dragged	O
along	O
in	O
the	O
maelstrom	O
for	O
a	O
while	O
,	O
but	O
when	O
Linux	B-Operating_System
came	O
along	O
,	O
I	O
was	O
actually	O
relieved	O
that	O
I	O
could	O
go	O
back	O
to	O
professoring	O
.	O
</s>
<s>
While	O
writing	O
MINIX	B-Operating_System
was	O
fun	O
,	O
I	O
do	O
n't	O
really	O
regard	O
it	O
as	O
the	O
most	O
important	O
thing	O
I	O
have	O
ever	O
done	O
.	O
</s>
<s>
While	O
Tanenbaum	O
himself	O
has	O
mentioned	O
that	O
he	O
did	O
not	O
write	O
the	O
article	O
to	O
renew	O
the	O
debate	O
on	O
kernel	B-Operating_System
design	I-Operating_System
,	O
</s>
<s>
Torvalds	O
posted	O
a	O
rebuttal	O
of	O
Tanenbaum	O
's	O
arguments	O
via	O
an	O
online	O
discussion	B-Application
forum	I-Application
,	O
</s>
<s>
This	O
prompted	O
Jonathan	O
Shapiro	O
(	O
primary	O
developer	O
of	O
the	O
EROS	B-Operating_System
microkernel	I-Operating_System
)	O
to	O
respond	O
that	O
most	O
of	O
the	O
field-proven	O
reliable	O
and	O
secure	O
computer	O
systems	O
use	O
a	O
more	O
microkernel-like	O
approach	O
.	O
</s>
<s>
In	O
2017	O
,	O
it	O
was	O
revealed	O
that	O
Intel	O
was	O
running	O
MINIX	B-Operating_System
inside	O
the	O
Management	B-Device
Engine	I-Device
,	O
a	O
separate	O
processor	O
incorporated	O
in	O
Intel	O
processor	O
chipsets	O
since	O
2008	O
.	O
</s>
<s>
As	O
a	O
response	O
to	O
this	O
,	O
Tanenbaum	O
wrote	O
an	O
open	O
letter	O
to	O
Intel	O
claiming	O
MINIX	B-Operating_System
to	O
be	O
“	O
the	O
most	O
widely	O
used	O
computer	O
operating	O
system	O
in	O
the	O
world	O
”	O
.	O
</s>
<s>
The	O
letter	O
was	O
interpreted	O
by	O
the	O
community	O
that	O
Tanenbaum	O
considered	O
himself	O
to	O
have	O
finally	O
won	O
the	O
debate	O
,	O
even	O
though	O
the	O
number	O
of	O
systems	O
running	O
Linux	B-Operating_System
was	O
considered	O
to	O
outnumber	O
the	O
number	O
of	O
desktop	O
systems	O
with	O
Intel	O
chipsets	O
,	O
when	O
taking	O
embedded	O
systems	O
and	O
handheld	O
devices	O
in	O
consideration	O
.	O
</s>
<s>
As	O
the	O
revelations	O
regarding	O
the	O
Management	B-Device
Engine	I-Device
were	O
highly	O
controversial	O
with	O
security	O
exploits	O
and	O
back	O
doors	O
it	O
was	O
also	O
not	O
considered	O
as	O
a	O
particularly	O
flattering	O
achievement	O
.	O
</s>
<s>
It	O
was	O
also	O
condemned	O
as	O
being	O
out	O
of	O
touch	O
as	O
the	O
only	O
criticism	O
Tanenbaum	O
laid	O
out	O
towards	O
Intel	O
was	O
that	O
he	O
was	O
never	O
informed	O
of	O
the	O
use	O
of	O
MINIX	B-Operating_System
in	O
the	O
Intel	O
chipsets	O
.	O
</s>
