Re: [9fans] EPIA dual-processor motherboard runs Plan 9

Re: [9fans] EPIA dual-processor motherboard runs Plan 9

[Richard Miller]

> depends what's mapped at those addresses. i7 is likely
> an apic or somesuch and shouldn't be a problem. i6 is
> just below 512MB physical and might be a problem depending
> on how much memory is in the machine; the e820 map
> would be useful here, maybe it's some acpi thing.

E820: 00000000 0009f000 memory
E820: 0009f000 000a0000 reserved
E820: 000f0000 00100000 reserved
E820: 00100000 1bde0000 memory
E820: 1bde0000 1bde3000 acpi nvs
E820: 1bde3000 1bdf0000 acpi reclaim
E820: 1bdf0000 1be00000 reserved
E820: fec00000 100000000 reserved

There's 512MB of physical RAM, but the top 64M is shared with
onboard VGA.

Re: [9fans] MS Research reinvents Inferno?

[Brian L.Stuart]

> I've several texts on OS that are pretty good for undergrad class, but
> is there an equivalent to Knuth's `Art of Computer Programming' or the
> second Dragon book?  Seems that a lot of wasted effort could be
> avoided by having a good servey of the ideas and references to
> previous work.

My first reaction is to say be patient, I just signed the contract a couple
of weeks ago.  My other reaction is to say I can't presume to be compared
to such august company.  But the reality is that I am in the process of
writing an OS textbook with the intent of making a small step in that
direction.

The rest of this is details about the book, so feel free to move on to
the next message without hurting my feelings...
The book is divided into groups of three chapters, each group covering
a major topic.  The major topics are Intro (history, organization, etc),
processes, memory, I/O, file systems, security and distributed systems.
There's also a group of three appendices covering a review of hardware/
architecture.  The first chapter in each group covers the general principles
and gives a brief view into how that topic is implemented in a number of
example systems.  The exact set of examples is not finalized but will at
least include MULTICS, RT-11, 6th edition UNIX, 4.3BSD, VMS, NT/2000/XP
and Xen.  Most likely Plan 9 will also appear there.  I want to also include
at least one example from the big iron world, but need to find a good
reference for it.  The second and third chapters of each section take
a detailed look at implementation.  The second chapter is on Linux and
the third on Inferno.  The plan is for these chpaters to cover a level of
detail something like Lions did for 6th edition UNIX, but probably not
quite as thorough.  Rather than separate the code and commentary
like Lions did, I'm making it look more similar to Knuth's literate programming
where the commentary is interspersed with pretty-printed code.

If we manage to keep to the publisher's schedule and I survive the process,
it should see the light of day in time for fall semester 2007.  So if you hear
the murmer of insane babbling coming from the direction of Memphis,
you know what it is.

Brian L. Stuart

Re: [9fans] MS Research reinvents Inferno?

[Paweł Lasek]

[snip]
>  The exact set of examples is not finalized but will at
> least include MULTICS, RT-11, 6th edition UNIX, 4.3BSD, VMS, NT/2000/XP
> and Xen.  Most likely Plan 9 will also appear there.  I want to also include
> at least one example from the big iron world, but need to find a good
> reference for it.

Hercules emulator allows you to run virtually any OS for S/390 and descendants,
and there are pre-made images somewhere on the Net with turn-key IBM
MVS installation.

http://www.cray-cyber.org provides public supercomputer systems, from
Cray, Control Data Corporation, NEC and Bull :) (guest account is down
due to an attack where it was used -_- )

The OS'es there are:
Cray: UNICOS(TM) - Unix based
CDC 6600-compatibles: NOS or NOS/VE
CDC 4680: EP/IX (UNIX based)
NEC SX-4: Super-UX - Unix based
NEC UP4800: UX4800 11.5

Pity is that only Cray Y-MP EL is 24/7, so you may have to talk with
them to have access to others

> If we manage to keep to the publisher's schedule and I survive the process,
> it should see the light of day in time for fall semester 2007.  So if you hear
> the murmer of insane babbling coming from the direction of Memphis,
> you know what it is.

> Brian L. Stuart

Keep up with good work, my "Operating Systems Vade Mecum" (R.A.Finkel)
is getting a little old :)

--
Paweł Lasek
"Once a hitokiri, always a hitokiri. This will never change" - Jine-Ei
http://plasek.jogger.pl [in polish]
http://plasek.wordpress.com [in polish]

[9fans] EPIA dual-processor motherboard runs Plan 9

[Richard Miller]

Fans of the (tiny, quiet, low-powered) VIA EPIA mini-itx motherboards
may be interested to know that the new VT-310DP with 2 x 1Ghz Eden-N
processors works very well with Plan 9.

Selected boot messages:

Plan 9
...
cpu0: 998MHz CentaurHauls unknown (cpuid: AX 0x069A DX 0x381BA3F)
ELCR: 0E20
LAPIC: fee00000 e0000000
cpu1: 997MHz CentaurHauls unknown (cpuid: AX 0x069A DX 0x381BA3F)
mtrrfix1: i9: 404040404040404 0
mtrrfix1: i10: 404040404040400 0
mtrrvar1: i6: 1BE00000 0
mtrrvar1: i7: FFFE00800 0
pcirouting: PCI.0.15.0 at pin 1 link 1 irq 11
#l0: i82557: 100Mbps port 0xE100 irq 10: 00e081xxxxxx
#l1: vt6102: 100Mbps port 0xD800 irq 11: 00e081xxxxxx
#U/usb0: uhci: port 0xDD00 irq 11
#U/usb1: uhci: port 0xDE00 irq 11
#U/usb2: uhci: port 0xDF00 irq 10
#U/usb3: uhci: port 0xE000 irq 10

Does anyone know whether I should be worried about the mtrr messages?

Output of pci command:

0.0.0:	---  06.00.00 1106/0259   0
0.0.1:	---  06.00.00 1106/1259   0
0.0.2:	---  06.00.00 1106/2259   0
0.0.3:	---  06.00.00 1106/3259   0
0.0.4:	---  06.00.00 1106/4259   0
0.0.7:	---  06.00.00 1106/7259   0
0.1.0:	---  06.04.00 1106/b198   0
0.10.0:	net  02.00.00 1106/3119   5 0:0000d401 256 1:f7041000 256
0.15.0:	disk 01.01.8a 1106/0571  11 4:0000dc01 16
0.16.0:	---  0c.03.00 1106/3038  11 4:0000dd01 32
0.16.1:	---  0c.03.00 1106/3038  11 4:0000de01 32
0.16.2:	---  0c.03.00 1106/3038  10 4:0000df01 32
0.16.3:	---  0c.03.00 1106/3038  10 4:0000e001 32
0.17.0:	---  06.01.00 1106/3227   0
0.18.0:	net  02.00.00 1106/3065  11 0:0000d801 256 1:f7043000 256
0.8.0:	vid  03.00.00 1002/4750 255 0:f6000008 16777216 1:0000d001 256 2:f7040000 4096
0.9.0:	net  02.00.00 8086/1229  10 0:f7044000 4096 1:0000e101 64 2:f7000000 131072
1.0.0:	vid  03.00.00 1106/3118  11 0:f0000008 67108864 1:f4000000 16777216

Other notes:

- Two of the three (!) onboard ethernet ports are supported; the third is
  gigabit ether and Plan 9 doesn't see it
- Native VGA just about works in vesa mode (1024x768 only; no acceleration;
  aux/vga -p sometimes hangs), but there's a PCI slot so you can add a VGA card

-- Richard

Re: [9fans] EPIA dual-processor motherboard runs Plan 9

[Ronald G Minnich]

Richard Miller wrote:
> Fans of the (tiny, quiet, low-powered) VIA EPIA mini-itx motherboards
> may be interested to know that the new VT-310DP with 2 x 1Ghz Eden-N
> processors works very well with Plan 9.
> 
> Selected boot messages:
> 
> Plan 9
> ...
> cpu0: 998MHz CentaurHauls unknown (cpuid: AX 0x069A DX 0x381BA3F)
> ELCR: 0E20
> LAPIC: fee00000 e0000000
> cpu1: 997MHz CentaurHauls unknown (cpuid: AX 0x069A DX 0x381BA3F)
> mtrrfix1: i9: 404040404040404 0
> mtrrfix1: i10: 404040404040400 0

"fixed" as in fixed-address-space-size MTRRs, not as in "they used to be 
broken" -- these look right to me IIRC.

> mtrrvar1: i6: 1BE00000 0
> mtrrvar1: i7: FFFE00800 0

var -- as in "variable sized address space" MTRRs.

These are familiar values, to me as well. I have not looked them up, but 
they sure sound like values I am used to. I don't think you need to worry.

ron

Re: [9fans] EPIA dual-processor motherboard runs Plan 9

[jmk]

On Sat Dec 31 10:48:44 EST 2005, rminnich@lanl.gov wrote:
> Richard Miller wrote:
> ...
> > mtrrvar1: i6: 1BE00000 0
> > mtrrvar1: i7: FFFE00800 0
> 
> var -- as in "variable sized address space" MTRRs.
> 
> These are familiar values, to me as well. I have not looked them up, but 
> they sure sound like values I am used to. I don't think you need to worry.
> 
> ron

the concern here is that the mttr values are different
on the 2 processors so they are seeing different memory
attributes for those ranges (the warning messages are
only printed if there is an inconsistency).

depends what's mapped at those addresses. i7 is likely
an apic or somesuch and shouldn't be a problem. i6 is
just below 512MB physical and might be a problem depending
on how much memory is in the machine; the e820 map
would be useful here, maybe it's some acpi thing.

"fixing" the mttrs as in making them consistent would be
a pain.

--jim

Re: [9fans] EPIA dual-processor motherboard runs Plan 9

[jmk]

sorry, i ran over my 'r' quota in the last message
and had to say 'mttr' instead of 'mtrr' everywhere.

Re: [9fans] MS Research reinvents Inferno?

[David Leimbach]

On 12/14/05, C H Forsyth <forsyth@vitanuova.com> wrote:
> speaking of OS textbooks and making a fuss about not-too-original work,
> surely the best recent example is the much-touted O(1) linux scheduler!
> was it not as many as 10 or 12 years waiting for someone, somewhere to read
> one of the above mentioned textbooks ...
>
>

Not sure I understand.  I've read about how this scheduler works, it
was so simple sounding at the time I've forgotten now :).

Are you saying that this was discussed in Computer Science texts 10
years ago and no one implemented it?

I'm pretty sure that's par for the course.  There's far too much "if
it ain't broke..." mentality that people never strive to create things
that are just flat out *better*.

Re: [9fans] MS Research reinvents Inferno?

[Brian L.Stuart]

> Hercules emulator allows you to run virtually any OS for S/390 and descendants,
> and there are pre-made images somewhere on the Net with turn-key IBM
> MVS installation.
> http://www.cray-cyber.org provides public supercomputer systems, from
> Cray, Control Data Corporation, NEC and Bull :) (guest account is down
> due to an attack where it was used -_- )

Thanks for the suggestions.  I'll look into them.  I've thought that MVS
might make a good big iron example, but I've got to admit that I don't
know much about its internals.  I have a feeling that I'll end up adding
as many examples as I have time for.  The unfortuate part is that the
schedule doesn't look to have time in abundance.

Brian L. Stuart

Re: [9fans] MS Research reinvents Inferno?

[Brian L.Stuart]

> speaking of OS textbooks and making a fuss about not-too-original work,
> surely the best recent example is the much-touted O(1) linux scheduler!
> was it not as many as 10 or 12 years waiting for someone, somewhere to read
> one of the above mentioned textbooks ...

Good point.  I'll make sure I reference earlier descriptions when I talk
about the Linux scheduler.

Brian L. Stuart

Re: [9fans] MS Research reinvents Inferno?

[C H Forsyth]

speaking of OS textbooks and making a fuss about not-too-original work,
surely the best recent example is the much-touted O(1) linux scheduler!
was it not as many as 10 or 12 years waiting for someone, somewhere to read
one of the above mentioned textbooks ...

Re: [9fans] MS Research reinvents Inferno?

[Brantley Coile]

I've several texts on OS that are pretty good for undergrad class, but
is there an equivalent to Knuth's `Art of Computer Programming' or the
second Dragon book?  Seems that a lot of wasted effort could be
avoided by having a good servey of the ideas and references to
previous work.  Bell & Newell, and Blaauw & Brooks, did this for
Architectures.  A journey thru the Dragon book's bibliography is like
a time machine to learn from the past.

Re: [9fans] MS Research reinvents Inferno?

[Brantley Coile]

>  - nice to see forward slashes as path delimiters in a Microsoft OS
> (other than Xenix)

When a company tries to buy up all the talent there are some
disadvantages.  You can't make sure everyone thinks Tenex was
the best OS ever. :)

Re: [9fans] MS Research reinvents Inferno?

[erik quanstrom]

this isn't really about the kernel, but  i think there are two important
bug-reducers in plan9 that windows and *nix systems don't have:

1. plan9 picks a standard and sticks with it. the best example is character set.
plan9 picks one and implements it. and things just work. gnu grep doesn't
and it has many character-set related bugs. for example, in a utf-8 locale,
it's about 100x slower than in a us-ascii locale.

2. plan9's utf-8/unicode implementation also avoids bugs by not
implementing bidi, character shaping, combining characters or extended planes.
to which most of the code in full unicode implentations is devoted 

- erik

Russ Cox <rsc@swtch.com> writes

| 
| > It would be interesting to see a bugs-per-million-lines-of-code
| > statistic for Plan 9 vs other systems.  In my experience, Plan 9's
| > advantage is that it's small and simple enough that the bugs are
| > relatively easy to find and fix.
| 
| I agree.  I started toward figuring that out for Plan 9 by
| putting together the kernel history at http://swtch.com/plan9history.
| Step 2 is understanding every change and categorizing them,
| including whether or not they are bug fixes.  Then one can derive
| many interesting statistics about the bugs.
| 
| I got about a quarter of the way through categorizing
| http://swtch.com/cgi-bin/plan9history.cgi?p=^(pc|port)/.*\.[ch]&v=filelist
| I'm hoping to find a student at MIT to finish the job.
| 
| Russ

Re: [9fans] MS Research reinvents Inferno?

[LiteStar numnums]

[-- Attachment #1: Type: text/plain, Size: 1162 bytes --]

Well, if we're going to be prickly, we could also include Services for Unix,
which is just a retargeted OpenBSD userspace.

On 12/14/05, Jack Johnson <knapjack@gmail.com> wrote:
>
> On 12/13/05, LiteStar numnums <litestar@gmail.com> wrote:
> > Let's see, first there was the Cyclone clone (Vault C), now there is a
> > Plan9-ish/EROS-ish kernel...
> >  Before all of this was "Active Directory" (Kerberos/LDAP)...
>
> You forgot Hydra (Citrix). And DOS (CP/M).
>
> -Jack
>



--
Nietzsche's first step is to accept what he knows. Atheism for him goes
without saying and is "contructive and
radical". Nietzsche's supreme vocation, so he says, is to provoke a kind of
crisis and a final decision about the
problem of atheism. The world continues on its course at random and there is
nothing final about it. Thus God
is useless, since He wants nothing in particular. If he wanted something --
and here we recognize the traditional
forumlation of the problem of evil -- He would have to assume responsiblity
for "a sum total of pain and inconsistency
which would debase the entire value of being born."
-- Albert Camus, L'Homme révolté

[-- Attachment #2: Type: text/html, Size: 1560 bytes --]

Re: [9fans] MS Research reinvents Inferno?

[Russ Cox]

> It would be interesting to see a bugs-per-million-lines-of-code
> statistic for Plan 9 vs other systems.  In my experience, Plan 9's
> advantage is that it's small and simple enough that the bugs are
> relatively easy to find and fix.

I agree.  I started toward figuring that out for Plan 9 by
putting together the kernel history at http://swtch.com/plan9history.
Step 2 is understanding every change and categorizing them,
including whether or not they are bug fixes.  Then one can derive
many interesting statistics about the bugs.

I got about a quarter of the way through categorizing
http://swtch.com/cgi-bin/plan9history.cgi?p=^(pc|port)/.*\.[ch]&v=filelist
I'm hoping to find a student at MIT to finish the job.

Russ

Re: [9fans] MS Research reinvents Inferno?

[Richard Miller]

> Inferno and Plan 9 are reliable
> mainly because they don't have many bugs.

It would be interesting to see a bugs-per-million-lines-of-code
statistic for Plan 9 vs other systems.  In my experience, Plan 9's
advantage is that it's small and simple enough that the bugs are
relatively easy to find and fix.

-- Richard

Re: [9fans] MS Research reinvents Inferno?

[C H Forsyth]

> But there's no checking of the sequencing of communications (e.g. you
> can't specify that SYN in one direction is followed by ACK in the other).

there was a variant that did more than that, using a protocol description
based on regular expressions, by the late A J Fisher (University of York)
it included compile-time checking.

	Fisher A J 1988  A critique of OCCAM channel types.  Computer Languages 13(2) 95-105

it's also discussed in another article to supply a theoretical basis

	Carlisle W H, Type Checking Concurrent I/O, TOPLAS 17(3) 445-469

Re: [9fans] MS Research reinvents Inferno?

[Richard Miller]

> Maybe I'm just sheltered, but I've never seen seconds-long pauses in a
> GC'd system.  Not even a bad Java app.

I've just ordered some more RAM for my thinkpad in the hope of alleviating
the seconds-long pauses I see when running eclipse.

Re: [9fans] MS Research reinvents Inferno?

[Richard Miller]

>> So... then there is no obvious reason, not to use Plan9. ;-)
> 
> no web browser :P

mothra - /n/sources/extra/mothra
charon - http://www.vitanuova.com/inferno/downloads.html

-- Richard

Re: [9fans] MS Research reinvents Inferno?

[Steve Simon]

> >  - are there symbolic links in any other Microsoft OSes?
> 
> Shortcuts resemble symbolic links.  I never looked any deeper than that.

Danger Will Robinson!

Off topic I know but I have recently spent some heartache
understanding shortcuts on windows:

Shortcuts are a Windows explorer feature, you can click through
a shortcut and windows open-a-file dialogue then the application gets
the resolved windows path, however you cannot use shortcuts in
console applications.

If your Windows OS of choice sits on top of NTFS then you can create
the equivilent of hard links (Reparse points in MS parlence) which
exist in the disk driver so all programs obey them. There are some
tools to manipulate these on MS's website.

-Steve

Re: [9fans] MS Research reinvents Inferno?

[Richard Miller]

>> i thought that was potentially interesting (and there were other things), but i was
>> also fairly sure i'd seen that done for (a variant of) occam-2
> 
> I thought I was taught occam channels where always typed
> and the message types where fixed - though this was quite a while ago.

Occam 2 channel typing was extended with what was called 'protocols',
but the term was a bit misleading.  It was just a way of declaring the
channel to pass a complex type (e.g. a counted array, or a discriminated
union - what limbo calls an adt with a pick clause).   The compiler
checks the correctness of every individual send/receive (e.g. that the
right number of elements are communicated for a counted array - no
buffer overflows! - and that all cases of a discriminated union are handled).
But there's no checking of the sequencing of communications (e.g. you
can't specify that SYN in one direction is followed by ACK in the other).

> I was sad Transputers & occam missed.

In fact there are huge numbers of transputers today hidden inside
consumer devices, programmed presumably in occam because that's
essentially their machine language - they just aren't identified as such.

-- Richard

Re: [9fans] MS Research reinvents Inferno?

[Russ Cox]

> /dev = /dev
> ServiceProviderContract = "mount"
> /hardware/drivers subtree = "pound (#)" devices (section 3 of the manual)

And you can't bind '#p' onto /dev because you'll get
a type error.

Here's another fun one: bind /net/tcp /proc; ps

Russ

RE: [9fans] MS Research reinvents Inferno?

[Andriy G. Tereshchenko]

 Jack Johnson wrote: 
>  - are there symbolic links in any other Microsoft OSes?

WinNT does use symbolic links and have global namespace internaly.
You take a look on it here http://www.sysinternals.com/Utilities/WinObj.html

Re: [9fans] MS Research reinvents Inferno?

[Jack Johnson]

Another interesting tidbit:

"Language runtimes provide services—notably garbage collection—that
can interact poorly with programs. For example, a generational garbage
collector may introduce seconds-long pauses in program execution,
which would disrupt a media player or operating system. On the other
hand, a real-time collector suitable for the media player might
penalize a computational task."

Maybe I'm just sheltered, but I've never seen seconds-long pauses in a
GC'd system.  Not even a bad Java app.

"Homogeneous environments also evolve into large, complex, and
expensive systems since they must support the union of the
requirements of every application that depends on them."

This also seems a little biased. I think there are excellent examples
of simplicity run amok in various operating systems that help prevent
devolution into overt complexity, including the OS presented.  I think
GC in general gets a bad rap, and coming up with a systemwide GC
policy might have been a good thing.

But now I'm showing my bias....

-Jack

Re: [9fans] MS Research reinvents Inferno?

[Jack Johnson]

On 12/13/05, Russ Cox <rsc@swtch.com> wrote:
> Perhaps they should have mentioned Oberon, but they did reference
> five other systems written in checked languages compiled to binary,
> indistinguishable from Oberon as far as this point is concerned.

True, and they do mention Cedar, which was Wirth's inspiration for
Oberon, so I suspect they made every attempt to cover their bases.

They probably took more inspiration through reflection (no pun
intended) than direct inspection, anyway.

-Jack

Re: [9fans] MS Research reinvents Inferno?

[andrey mirtchovski]

> "poung (#)"

heh, sorry. had too much cyrillic tonight.

Re: [9fans] MS Research reinvents Inferno?

[andrey mirtchovski]

> "Finally, the /dev namespace is a public directory holding symbolic links to
> ServiceProviderContract endpoints in the /hardware/drivers subtree. In
> this manner,
> an application can be bound to a public name, without knowing the true
> name of the driver."

i'll translate:

/dev = /dev
ServiceProviderContract = "mount"
/hardware/drivers subtree = "poung (#)" devices (section 3 of the manual)

:)

Re: [9fans] MS Research reinvents Inferno?

[Lucio De Re]

>  - are there symbolic links in any other Microsoft OSes?

Shortcuts resemble symbolic links.  I never looked any deeper than that.

++L

Re: [9fans] MS Research reinvents Inferno?

[Jack Johnson]

On 12/13/05, Russ Cox <rsc@swtch.com> wrote:
> Apologies to those who did read the paper.

Ah, we can still be opinionated bastards anyway. :)

Trivia, from the paper:

"Finally, the /dev namespace is a public directory holding symbolic links to
ServiceProviderContract endpoints in the /hardware/drivers subtree. In
this manner,
an application can be bound to a public name, without knowing the true
name of the driver."

Just randomly caught my eye for three reasons:

 - /dev seems awfully Unixish
 - nice to see forward slashes as path delimiters in a Microsoft OS
(other than Xenix)
 - are there symbolic links in any other Microsoft OSes?

-Jack

Re: [9fans] MS Research reinvents Inferno?

[Jack Johnson]

On 12/13/05, LiteStar numnums <litestar@gmail.com> wrote:
> Let's see, first there was the Cyclone clone (Vault C), now there is a
> Plan9-ish/EROS-ish kernel...
>  Before all of this was "Active Directory" (Kerberos/LDAP)...

You forgot Hydra (Citrix). And DOS (CP/M).

-Jack

Re: [9fans] MS Research reinvents Inferno?

[Bruce Ellis]

Both ozinferno and guido do bounds checking.  It's not hard.
It chews a few cycles under emu but for guido it's just a
bucket of gates.

brucee

On 12/14/05, Charles Forsyth <forsyth@terzarima.net> wrote:
> > when Inferno is jitting, it doesn't
> > insert bounds checks on array references, so it can crash easily too.
>
> as it happens, we've added those checks.  ok, ok, they're currently wierdly optional,
> and i'd need to check which jits have had them added so far,
> but probably the default will change once we've a better idea of the cost.
> it saves a lot of unpleasantness.

Re: [9fans] MS Research reinvents Inferno?

[erik quanstrom]

i don't think it's possible to like or dislike ibm or m$ as a whole.
they're much too big for a monolitic opinion

- erik

Ronald G Minnich <rminnich@lanl.gov> writes

| 
| David Leimbach wrote:
| 
| > I'd probably get some of those "new" Intel/AMD CPUs that will have 
| > integrated hypervisor support and work on something that's more like an 
| > Exokernel/Xen hybrid.
| 
| worth noting that the xbox does this. It has a hypervisor. It is a very 
| innovative piece of gear in many ways, worth learning from. The Xbox 
| (and probably the ps/3) are going to bring hypervisor to the masses -- 
| and the hypervisor smarts are at IBM.
| 
| There's some neat stuff going on, even if you might not like the folks 
| doing it.
| 
| ron

Re: [9fans] MS Research reinvents Inferno?

[andrey mirtchovski]

> So... then there is no obvious reason, not to use Plan9. ;-)

no web browser :P

Re: [9fans] MS Research reinvents Inferno?

[Oliver Bandel]

On Tue, Dec 13, 2005 at 12:52:35PM -0800, David Leimbach wrote:
> On 12/13/05, Oliver Bandel <oliver@first.in-berlin.de> wrote:
> >
> > On Tue, Dec 13, 2005 at 05:56:19PM +0000, Charles Forsyth wrote:
> > > > when Inferno is jitting, it doesn't
> > > > insert bounds checks on array references, so it can crash easily too.
> > >
> > > as it happens, we've added those checks.  ok, ok, they're currently
> > wierdly optional,
> > > and i'd need to check which jits have had them added so far,
> > > but probably the default will change once we've a better idea of the
> > cost.
> > > it saves a lot of unpleasantness.
> > >
> >
> >
> > What's about Ocaml-ports to Plan9?
> >
> > Would be fine to have it as a development tool. :)
> > (IMHO it's necessary to have it on each platform.)
> >
> >
> > Ciao,
> >    Oliver
> 
> 
> 
> Use google.. I think Andrey did this already
[...]

OK, I didn't expected it, so I didn't looked for it.

But it's good that this job was done. :)

So... then there is no obvious reason, not to use Plan9. ;-)

(Are there non-obvious ones?)

Ciao,
   Oliver

Re: [9fans] MS Research reinvents Inferno?

[Oliver Bandel]

On Tue, Dec 13, 2005 at 04:05:04PM -0500, Brantley Coile wrote:
> > yeah but then you'd have to write in Modula3
> 
> Can't.  Ref man > 50 pages.  Isn't there some rule about
> languages that take more than 50 pages to define?
> 

who reads such pages? ;-)

Re: [9fans] MS Research reinvents Inferno?

[Steve Simon]

On Tue Dec 13 17:57:37 GMT 2005, forsyth@terzarima.net wrote:
> > They push the type-checking further in other areas too.
> > The channels are two-way, with defined protocols running over
> > them.  Each reference to a channel is marked as to which half
> > of the protocol it is expected to run, and then the code using the
> 
> i thought that was potentially interesting (and there were other things), but i was
> also fairly sure i'd seen that done for (a variant of) occam-2

I thought I was taught occam channels where always typed
and the message types where fixed - though this was quite a while ago.

What always amazed me was the debuggers ability to push through
a channel to another thread on a different cpu.

I was sad Transputers & occam missed.

-Steve

Re: [9fans] MS Research reinvents Inferno?

[Skip Tavakkolian]

> The subject of this thread is about MS Research reinventing Inferno... I
> think that's a good indication some people thought it was "merely"
> derivative :)

it was a question, hence the '?'

my point, as i clarified later, is the lack of acknowledgement for
prior work.  the victors write the history.  how the majority
perceive the history, becomes reality.

btw, the first i saw the concept of combining resources, services and
files in a name space, was plan9.  am i wrong?

Re: [9fans] MS Research reinvents Inferno?

[David Leimbach]

[-- Attachment #1: Type: text/plain, Size: 802 bytes --]

On 12/13/05, Skip Tavakkolian <9nut@9netics.com> wrote:
>
> > The subject of this thread is about MS Research reinventing Inferno... I
> > think that's a good indication some people thought it was "merely"
> > derivative :)
>
> it was a question, hence the '?'
>
> my point, as i clarified later, is the lack of acknowledgement for
> prior work.  the victors write the history.  how the majority
> perceive the history, becomes reality.




btw, the first i saw the concept of combining resources, services and
> files in a name space, was plan9.  am i wrong?
>
>

I'm not saying you're wrong.  I'm saying that I don't think many ideas are
truly "original" to begin with.  Just combinations of past ideas in new ways
perhaps.

Wish we could convince the patent system of that.

[-- Attachment #2: Type: text/html, Size: 1313 bytes --]

Re: [9fans] MS Research reinvents Inferno?

[LiteStar numnums]

[-- Attachment #1: Type: text/plain, Size: 1501 bytes --]

I would think that OpenVMS had the ruddimentary ideas, as well as several
other OSs of the time (MUMPS did a semi-decent job back in the day...).
Obviously, plan9 pulled the ideas forward & unified them better, but it
wasn't wholly original.

On 12/13/05, Skip Tavakkolian <9nut@9netics.com> wrote:
>
> > The subject of this thread is about MS Research reinventing Inferno... I
> > think that's a good indication some people thought it was "merely"
> > derivative :)
>
> it was a question, hence the '?'
>
> my point, as i clarified later, is the lack of acknowledgement for
> prior work.  the victors write the history.  how the majority
> perceive the history, becomes reality.
>
> btw, the first i saw the concept of combining resources, services and
> files in a name space, was plan9.  am i wrong?
>
>


--
Nietzsche's first step is to accept what he knows. Atheism for him goes
without saying and is "contructive and
radical". Nietzsche's supreme vocation, so he says, is to provoke a kind of
crisis and a final decision about the
problem of atheism. The world continues on its course at random and there is
nothing final about it. Thus God
is useless, since He wants nothing in particular. If he wanted something --
and here we recognize the traditional
forumlation of the problem of evil -- He would have to assume responsiblity
for "a sum total of pain and inconsistency
which would debase the entire value of being born."
-- Albert Camus, L'Homme révolté

[-- Attachment #2: Type: text/html, Size: 1880 bytes --]

Re: [9fans] MS Research reinvents Inferno?

[Brantley Coile]

> yeah but then you'd have to write in Modula3

Can't.  Ref man > 50 pages.  Isn't there some rule about
languages that take more than 50 pages to define?

Re: [9fans] MS Research reinvents Inferno?

[David Leimbach]

[-- Attachment #1: Type: text/plain, Size: 828 bytes --]

On 12/13/05, Oliver Bandel <oliver@first.in-berlin.de> wrote:
>
> On Tue, Dec 13, 2005 at 05:56:19PM +0000, Charles Forsyth wrote:
> > > when Inferno is jitting, it doesn't
> > > insert bounds checks on array references, so it can crash easily too.
> >
> > as it happens, we've added those checks.  ok, ok, they're currently
> wierdly optional,
> > and i'd need to check which jits have had them added so far,
> > but probably the default will change once we've a better idea of the
> cost.
> > it saves a lot of unpleasantness.
> >
>
>
> What's about Ocaml-ports to Plan9?
>
> Would be fine to have it as a development tool. :)
> (IMHO it's necessary to have it on each platform.)
>
>
> Ciao,
>    Oliver



Use google.. I think Andrey did this already

http://pages.cpsc.ucalgary.ca/~mirtchov/p9/

[-- Attachment #2: Type: text/html, Size: 1239 bytes --]

Re: [9fans] MS Research reinvents Inferno?

[David Leimbach]

[-- Attachment #1: Type: text/plain, Size: 608 bytes --]

On 12/13/05, andrey mirtchovski <andrey@lanl.gov> wrote:
>
> >> That's the Oberon part.
> >
> > Some of that is true, but they do more checking than
> > Oberon, and they can verify the bytecodes they ship around
> > while Oberon could make no such guarantee about its binaries.
> >
> > Russ
>
> i thought the problem was that they didn't reference other systems. i
> don't think anyone claims their work is purely derivative.
>

The subject of this thread is about MS Research reinventing Inferno... I
think that's a good indication some people thought it was "merely"
derivative :)

Dave

[-- Attachment #2: Type: text/html, Size: 918 bytes --]

Re: [9fans] MS Research reinvents Inferno?

[David Leimbach]

[-- Attachment #1: Type: text/plain, Size: 590 bytes --]

On 12/13/05, William Josephson <jkw@eecs.harvard.edu> wrote:
>
> On Tue, Dec 13, 2005 at 12:43:27PM -0500, Russ Cox wrote:
> > Some of the kernel core (i.e., the low-level assembly, the garbage
> > collector, the debugger) is written in unchecked languages,
> > but most of it (including, for example, the scheduler and all the
> > device drivers) is written in checked languages.  Safe device drivers
> > alone would fix a huge fraction of the Windows crashes.
>
> To hear (some of) the UW guys tell it, it is ``just SPIN'' :-)
>

yeah but then you'd have to write in Modula3

[-- Attachment #2: Type: text/html, Size: 894 bytes --]

Re: [9fans] MS Research reinvents Inferno?

[Oliver Bandel]

On Tue, Dec 13, 2005 at 05:56:19PM +0000, Charles Forsyth wrote:
> > when Inferno is jitting, it doesn't
> > insert bounds checks on array references, so it can crash easily too.
> 
> as it happens, we've added those checks.  ok, ok, they're currently wierdly optional,
> and i'd need to check which jits have had them added so far,
> but probably the default will change once we've a better idea of the cost.
> it saves a lot of unpleasantness.
> 


What's about Ocaml-ports to Plan9?

Would be fine to have it as a development tool. :)
(IMHO it's necessary to have it on each platform.)


Ciao,
   Oliver

Re: [9fans] MS Research reinvents Inferno?

[Oliver Bandel]

On Tue, Dec 13, 2005 at 12:43:27PM -0500, Russ Cox wrote:
[...] 
> The real new research in Singularity is how far they are pushing
> type information into the deepest reaches of the system.

Maybe adapting OCaml to a F# (but IMHO some M$-people
also work on Haskell(?)) has opened the eyes to
such strong typed languages.


> 
> Some of the kernel core (i.e., the low-level assembly, the garbage
> collector, the debugger) is written in unchecked languages,
> but most of it (including, for example, the scheduler and all the
> device drivers) is written in checked languages.  Safe device drivers
> alone would fix a huge fraction of the Windows crashes.

[...]
> All this is toward the goal of reliability and dependability, as they
> clearly state in the introduction.  Inferno and Plan 9 are reliable
> mainly because they don't have many bugs.  Neither actually
> take steps to providing some form of safety guarantees.
> Plan 9 is running C code, and when Inferno is jitting, it doesn't
> insert bounds checks on array references, so it can crash easily too.

There is a small team of people (or was it a single developer?)
working on a Ocaml-bases OS.
So this means: type checking even inside the OS's kernel.

I lost the URl, but somehwere in my Mailfolders it should be
possible to find... (or google asking?).

Ciao,
   Oliver

Re: [9fans] MS Research reinvents Inferno?

[Oliver Bandel]

On Tue, Dec 13, 2005 at 11:26:20AM -0500, LiteStar numnums wrote:
> Let's see, first there was the Cyclone clone (Vault C), now there is a
> Plan9-ish/EROS-ish kernel...
> Before all of this was "Active Directory" (Kerberos/LDAP)...
> The 'research' seems to be focused upon many things already released into
> the wild &
> brining them to the point of being 'Microsoft' products...
[...]

as always done since decades...

... someone who knows F#?

It's a (not lossless) copy of Ocaml.
Ocaml lost OO-layer, but got .NET attached and then this is called
F#. ;-)

But there should be hundreds of other examples on how to get rich with
the idea of non-lossless copying by that company. ;-)

Ciao,
   Oliver

Re: [9fans] MS Research reinvents Inferno?

[Russ Cox]

> > Some of that is true, but they do more checking than
> > Oberon, and they can verify the bytecodes they ship around
> > while Oberon could make no such guarantee about its binaries.
> >
> > Russ
>
> i thought the problem was that they didn't reference other systems. i
> don't think anyone claims their work is purely derivative.

Perhaps they should have mentioned Oberon, but they did reference
five other systems written in checked languages compiled to binary,
indistinguishable from Oberon as far as this point is concerned.

Russ

Re: [9fans] MS Research reinvents Inferno?

[andrey mirtchovski]

>> That's the Oberon part.
>
> Some of that is true, but they do more checking than
> Oberon, and they can verify the bytecodes they ship around
> while Oberon could make no such guarantee about its binaries.
>
> Russ

i thought the problem was that they didn't reference other systems. i  
don't think anyone claims their work is purely derivative.

Re: [9fans] MS Research reinvents Inferno?

[Russ Cox]

> > Some of the kernel core (i.e., the low-level assembly, the garbage
> > collector, the debugger) is written in unchecked languages,
> > but most of it (including, for example, the scheduler and all the
> > device drivers) is written in checked languages.  Safe device drivers
> > alone would fix a huge fraction of the Windows crashes.
>
> That's the Oberon part.

Some of that is true, but they do more checking than
Oberon, and they can verify the bytecodes they ship around
while Oberon could make no such guarantee about its binaries.

Russ

Re: [9fans] MS Research reinvents Inferno?

[Brantley Coile]

> Some of the kernel core (i.e., the low-level assembly, the garbage
> collector, the debugger) is written in unchecked languages,
> but most of it (including, for example, the scheduler and all the
> device drivers) is written in checked languages.  Safe device drivers
> alone would fix a huge fraction of the Windows crashes.

That's the Oberon part.

Re: [9fans] MS Research reinvents Inferno?

[William Josephson]

On Tue, Dec 13, 2005 at 12:43:27PM -0500, Russ Cox wrote:
> Some of the kernel core (i.e., the low-level assembly, the garbage
> collector, the debugger) is written in unchecked languages,
> but most of it (including, for example, the scheduler and all the
> device drivers) is written in checked languages.  Safe device drivers
> alone would fix a huge fraction of the Windows crashes.

To hear (some of) the UW guys tell it, it is ``just SPIN'' :-)

Re: [9fans] MS Research reinvents Inferno?

[Ronald G Minnich]

Bakul Shah wrote:

> Has there been *any* revolutionary idea in CS since mid 1970s?
> upto 1975 (in no particular order)
>


all technologies go through these phases. The revolution happened. Now 
we're iterating on that solid foundation.

ron

Re: [9fans] MS Research reinvents Inferno?

[Ronald G Minnich]

David Leimbach wrote:

> I'd probably get some of those "new" Intel/AMD CPUs that will have 
> integrated hypervisor support and work on something that's more like an 
> Exokernel/Xen hybrid.

worth noting that the xbox does this. It has a hypervisor. It is a very 
innovative piece of gear in many ways, worth learning from. The Xbox 
(and probably the ps/3) are going to bring hypervisor to the masses -- 
and the hypervisor smarts are at IBM.

There's some neat stuff going on, even if you might not like the folks 
doing it.

ron

Re: [9fans] MS Research reinvents Inferno?

[alexandr babic]

Read parts: 5.2.5 Reflecting into the Namespace, 5.3 Nameserver, 5.4
Fileserver on pages 26, 27, 28 in the M$ pdf file:
ftp://ftp.research.microsoft.com/pub/tr/TR-2005-135.pdf

alex

Re: [9fans] MS Research reinvents Inferno?

[Bakul Shah]

> There are no real new ideas in CS popping around at this point, so we 
> are reduced to recycling each other's socks. So it goes. We're in an 
> evolutionary, not a revoluationary, business. This may be permanent, it 
> is hard to tell.

Has there been *any* revolutionary idea in CS since mid 1970s?
upto 1975 (in no particular order)
    Monitors
    critical regions
    synchronization primitives
    message passing
    rpc
    object oriented programming
    abstract data types
    garbage collection
    virtual memory
    virtual machines
    network file systems
    capabilities
    networking
    atomic transactions
    relational database
    denotational semantics
    functional programming
    parsing
    hashtables
    various search/sort techniques

the following may be post 1975 (but more likely I am mistaken)
    non blocking synchronization (more than just compare&swap)

My problem is not with reinvention [very few can be in the
right place at the right time with the right kind of brain]
but just how badly it is done.  Or that it starts out right
but then it is left incomplete.

Re: [9fans] MS Research reinvents Inferno?

[Charles Forsyth]

> They push the type-checking further in other areas too.
> The channels are two-way, with defined protocols running over
> them.  Each reference to a channel is marked as to which half
> of the protocol it is expected to run, and then the code using the

i thought that was potentially interesting (and there were other things), but i was
also fairly sure i'd seen that done for (a variant of) occam-2

Re: [9fans] MS Research reinvents Inferno?

[Charles Forsyth]

> when Inferno is jitting, it doesn't
> insert bounds checks on array references, so it can crash easily too.

as it happens, we've added those checks.  ok, ok, they're currently wierdly optional,
and i'd need to check which jits have had them added so far,
but probably the default will change once we've a better idea of the cost.
it saves a lot of unpleasantness.

Re: [9fans] MS Research reinvents Inferno?

[Skip Tavakkolian]

> I note that the paper didn't reference any other Plan 9'ish
> things other than Inferno.  Not having read the paper I assume
> that it is an Inferno like thing.  I wonder then, why they
> don't reference Wirth's and Gutknecht's Oberon system?
> Rob did.

That's the part that bugged me the most.  They obviously
know of prior effort. There are a lot of similarities; why don't they
want to admit that they're standing on the toes of giants. They
make one (back-handed) reference to Inferno, while there
are many similarities.

Is Ethics a required subject in Engineering anymore?

Re: [9fans] MS Research reinvents Inferno?

[Russ Cox]

Singularity is as much like Inferno as it is like a Java OS.
It ships around bytecodes as the stored form of programs,
and that's where the similarities end.  They don't jit the code,
instead treating it as a high-level program and compiling it
with a traditional optimizing compiler.

Yes, they have channels for communication, but that idea
predates Inferno and Plan 9 by quite a bit.  Also, the channels
are two-way with defined protocols running on them (not just
message types).

The real new research in Singularity is how far they are pushing
type information into the deepest reaches of the system.

Some of the kernel core (i.e., the low-level assembly, the garbage
collector, the debugger) is written in unchecked languages,
but most of it (including, for example, the scheduler and all the
device drivers) is written in checked languages.  Safe device drivers
alone would fix a huge fraction of the Windows crashes.

They push the type-checking further in other areas too.
The channels are two-way, with defined protocols running over
them.  Each reference to a channel is marked as to which half
of the protocol it is expected to run, and then the code using the
channel is verified to make sure it respects the protocol as far
as receiving or sending at the right points in time, sending only
the allowed set of messages, and handling all the possible
messages received.

All this is toward the goal of reliability and dependability, as they
clearly state in the introduction.  Inferno and Plan 9 are reliable
mainly because they don't have many bugs.  Neither actually
take steps to providing some form of safety guarantees.
Plan 9 is running C code, and when Inferno is jitting, it doesn't
insert bounds checks on array references, so it can crash easily too.

There is more, but I suggest you actually read the paper instead
of dismissing it by saying "Plan 9" or "Inferno" at first glance.
I find this to be perhaps the ugliest side of 9fans.

Apologies to those who did read the paper.
Russ

Re: [9fans] MS Research reinvents Inferno?

[Wes Kussmaul]

Ronald G Minnich wrote:

> There are no real new ideas in CS popping around at this point, so we 
> are reduced to recycling each other's socks. So it goes. We're in an 
> evolutionary, not a revoluationary, business. This may be permanent, it 
> is hard to tell.

Minutes ago I got a message from a journalist friend with just the 
subject _New OS Announced_ and a link:

http://www.prweb.com/releases/2005/12/prweb321987.htm

So ok, you needn't bother following that, it points to a press release 
by a white box pc vendor announcing that you can buy their computers 
with the new AX5 operating system that looks and works like Windows and 
opens MS Office files and runs this long list of Windows programs.

In other words it's some flavor of Linux with WINE and probably Open 
Office 2.0 and some hastily gimped graphics.

But to my friend it's a NEW OS!

Here's the thing. If someone has found a way to educate him to the fact 
that he can have these things and finds a way to get him using it, then 
indeed it IS a new OS.

So Ron, you say there's nothing new in CS, that we're just recycling 
each others' socks. How about combining socks with shirt and pants and 
offering a whole wardrobe? If that doesn't qualify as CS then so what? 
It's making computers more useful to people who don't know a lot about 
how they work.

With some effort and especially some imagination, someone could deliver 
something to my friend that not only emulates Windows but delivers much 
more than he ever knew he could have. How about a bounded online space 
where he can share files with his friends around the world with a fair 
degree of security? How about keeping DLL-land in its outdoor space, 
leaving it to its buffer overflows etc. but having an InDoor space a few 
keystrokes away?

I once read an article about companies selling specialized vehicles 
complaining about how their competitor had just put some special purpose 
equipment on a plain vanilla truck chassis and called it a specialized 
vehicle. A customer was then quoted saying how he preferred the latter 
not only because it was cheaper but that maintenance of the truck part 
was a commodity skill, parts were readily available etc. and it was just 
fine with him that someone found a way to quickly cobble something new 
out of readily available pieces. That, it seems to me, qualifies as 
truck science. But even if it doesn't it was probably a very fun thing 
to do. Sort of like finding a way to put Linux in ROM and replacing the 
stupid BIOS with it.


-- 
Wes Kussmaul
CIO
The Village Group
738 Main Street
Waltham, MA 02451

781-647-7178


My uncle likes to say that the world’s biggest troubles started when the 
serpent said, “Try this fruit, and by the way if a bunch of people 
collectively calling themselves Arthur Andersen signs something it’s the 
same as if a person named Arthur Andersen signed it.” I don’t get the 
serpent and fruit part. Must be some Swiss mythology thing. He can be a 
bit obscure.

                          P.K. Iggy
                          _How I Like Fixed The Internet_
                            (Tales from the Great Infodepression of 2009
                            and the prosperity that followed)

Re: [9fans] MS Research reinvents Inferno?

[David Leimbach]

[-- Attachment #1: Type: text/plain, Size: 1987 bytes --]

On 12/13/05, Ronald G Minnich <rminnich@lanl.gov> wrote:
>
> LiteStar numnums wrote:
> > Let's see, first there was the Cyclone clone (Vault C), now there is a
> > Plan9-ish/EROS-ish kernel...
>
> I did read the report, and IIRC they do give inferno its due.
>
> These are not dumb guys, I think Jim Laros at least deserves a little
> respect. Sure, it's M$, but at least it's not another bolt-on to the
> linux kernel, right?
>
> There are no real new ideas in CS popping around at this point, so we
> are reduced to recycling each other's socks. So it goes. We're in an
> evolutionary, not a revoluationary, business. This may be permanent, it
> is hard to tell.
>
> ron
>

I agree with Ron here.  I'm not often one to rush to Microsoft's aid.  They
are clearly pulling from ideas from L4, MUNGI, Inferno, SpinOS, and to some
extend even House [Haskell GHCI runtime used as an OS] etc...

They are combining them in a slightly different way, and exploring the
result.  I guess this still qualifies as research but it's not as
groundbreaking as some people would like.


If I were doing OS research today what would I do?

I'd probably get some of those "new" Intel/AMD CPUs that will have
integrated hypervisor support and work on something that's more like an
Exokernel/Xen hybrid.

Why "switch" OS images when you can just make the new hypervisor hardware
just switch out whole applications which have an even more convincing view
that they each own the whole machine.

Hell you could probably get good performance out of a Xen version of DOS
these days and do something interesting.  Yes, it's recycling ooooold stuff
to do this, but in a newer way.

It may not be as earth-shattering OS research as people might like to see,
but it's a hell of a lot better than just dorking around with Xwindows or
making yet another Unix clone.

So really, I'd just give em a break, at least they had the good taste to
look at Inferno :).

Dave

[-- Attachment #2: Type: text/html, Size: 2352 bytes --]

Re: [9fans] MS Research reinvents Inferno?

[alexandr babic]

> If you ask me, we need a fresh architecture.  Transputer, anyone?  Ah,
> that is where Transmeta got their name!  Undeservedly, it would seem.
> 
> ++L
> 

Today we use TRASHPUTER.

alex.

Re: [9fans] MS Research reinvents Inferno?

[Lucio De Re]

> There are no real new ideas in CS popping around at this point, so we 
> are reduced to recycling each other's socks. So it goes. We're in an 
> evolutionary, not a revoluationary, business. This may be permanent, it 
> is hard to tell.

If you ask me, we need a fresh architecture.  Transputer, anyone?  Ah,
that is where Transmeta got their name!  Undeservedly, it would seem.

++L

Re: [9fans] MS Research reinvents Inferno?

[Charles Forsyth]

> I note that the paper didn't reference any other Plan 9'ish
> things other than Inferno.  Not having read the paper I assume
> that it is an Inferno like thing.

not really; their reference to Inferno is a little misleading
(but that's not unusual for `Related Work').

Re: [9fans] MS Research reinvents Inferno?

[Ronald G Minnich]

LiteStar numnums wrote:
> Let's see, first there was the Cyclone clone (Vault C), now there is a 
> Plan9-ish/EROS-ish kernel...

I did read the report, and IIRC they do give inferno its due.

These are not dumb guys, I think Jim Laros at least deserves a little 
respect. Sure, it's M$, but at least it's not another bolt-on to the 
linux kernel, right?

There are no real new ideas in CS popping around at this point, so we 
are reduced to recycling each other's socks. So it goes. We're in an 
evolutionary, not a revoluationary, business. This may be permanent, it 
is hard to tell.

ron

Re: [9fans] MS Research reinvents Inferno?

[Brantley Coile]

[-- Attachment #1: Type: text/plain, Size: 251 bytes --]

I note that the paper didn't reference any other Plan 9'ish
things other than Inferno.  Not having read the paper I assume
that it is an Inferno like thing.  I wonder then, why they
don't reference Wirth's and Gutknecht's Oberon system?
Rob did.

[-- Attachment #2: Type: message/rfc822, Size: 7295 bytes --]

[-- Attachment #2.1.1: Type: text/plain, Size: 1612 bytes --]

Let's see, first there was the Cyclone clone (Vault C), now there is a
Plan9-ish/EROS-ish kernel...
Before all of this was "Active Directory" (Kerberos/LDAP)...
The 'research' seems to be focused upon many things already released into
the wild &
brining them to the point of being 'Microsoft' products...
 -- Stefan

On 12/13/05, Sergey Zhilkin <szhilkin@gmail.com> wrote:
>
> Microsoft reinventing the wheels since 1980 :)
> Any sources of Sing ?
> Inferno IS better :) and also OpenSource :)
>
> On 12/13/05, Skip Tavakkolian <9nut@9netics.com> wrote:
> > I'm not sure if this is old news to you all. It's something
> > that MS Research calls 'Singularity'.
> >
> > ftp://ftp.research.microsoft.com/pub/tr/TR-2005-135.pdf
> >
> > and a some interviews with the (re)inventors here:
> >
> > http://channel9.msdn.com/Showpost.aspx?postid=68302
> > http://channel9.msdn.com/Showpost.aspx?postid=141858
> >
> >
>



--
Nietzsche's first step is to accept what he knows. Atheism for him goes
without saying and is "contructive and
radical". Nietzsche's supreme vocation, so he says, is to provoke a kind of
crisis and a final decision about the
problem of atheism. The world continues on its course at random and there is
nothing final about it. Thus God
is useless, since He wants nothing in particular. If he wanted something --
and here we recognize the traditional
forumlation of the problem of evil -- He would have to assume responsiblity
for "a sum total of pain and inconsistency
which would debase the entire value of being born."
-- Albert Camus, L'Homme révolté

[-- Attachment #2.1.2: Type: text/html, Size: 2251 bytes --]

Re: [9fans] MS Research reinvents Inferno?

[Charles Forsyth]

> Yeah, but there's a key difference in that it isn't a product
> of the infinite monkey theorem.

between monkeys and emperor penguins, the o/s world
is obviously becoming a bit of a zoo.

Re: [9fans] MS Research reinvents Inferno?

[Paul Lalonde]

[-- Attachment #1: Type: text/plain, Size: 1770 bytes --]

Microsoft: Yesterday's Technology, Tomorrow.

On 13-Dec-05, at 8:26 AM, LiteStar numnums wrote:

> Let's see, first there was the Cyclone clone (Vault C), now there  
> is a Plan9-ish/EROS-ish kernel...
> Before all of this was "Active Directory" (Kerberos/LDAP)...
> The 'research' seems to be focused upon many things already  
> released into the wild &
> brining them to the point of being 'Microsoft' products...
>  -- Stefan
>
> On 12/13/05, Sergey Zhilkin <szhilkin@gmail.com> wrote:
> Microsoft reinventing the wheels since 1980 :)
> Any sources of Sing ?
> Inferno IS better :) and also OpenSource :)
>
> On 12/13/05, Skip Tavakkolian <9nut@9netics.com> wrote:
> > I'm not sure if this is old news to you all. It's something
> > that MS Research calls 'Singularity'.
> >
> > ftp://ftp.research.microsoft.com/pub/tr/TR-2005-135.pdf
> >
> > and a some interviews with the (re)inventors here:
> >
> > http://channel9.msdn.com/Showpost.aspx?postid=68302
> > http://channel9.msdn.com/Showpost.aspx?postid=141858
> >
> >
>
>
>
> -- 
> Nietzsche's first step is to accept what he knows. Atheism for him  
> goes without saying and is "contructive and
> radical". Nietzsche's supreme vocation, so he says, is to provoke a  
> kind of crisis and a final decision about the
> problem of atheism. The world continues on its course at random and  
> there is nothing final about it. Thus God
> is useless, since He wants nothing in particular. If he wanted  
> something -- and here we recognize the traditional
> forumlation of the problem of evil -- He would have to assume  
> responsiblity for "a sum total of pain and inconsistency
> which would debase the entire value of being born."
> -- Albert Camus, L'Homme révolté


[-- Attachment #2: Type: text/html, Size: 2532 bytes --]

Re: [9fans] MS Research reinvents Inferno?

[LiteStar numnums]

[-- Attachment #1: Type: text/plain, Size: 1612 bytes --]

Let's see, first there was the Cyclone clone (Vault C), now there is a
Plan9-ish/EROS-ish kernel...
Before all of this was "Active Directory" (Kerberos/LDAP)...
The 'research' seems to be focused upon many things already released into
the wild &
brining them to the point of being 'Microsoft' products...
 -- Stefan

On 12/13/05, Sergey Zhilkin <szhilkin@gmail.com> wrote:
>
> Microsoft reinventing the wheels since 1980 :)
> Any sources of Sing ?
> Inferno IS better :) and also OpenSource :)
>
> On 12/13/05, Skip Tavakkolian <9nut@9netics.com> wrote:
> > I'm not sure if this is old news to you all. It's something
> > that MS Research calls 'Singularity'.
> >
> > ftp://ftp.research.microsoft.com/pub/tr/TR-2005-135.pdf
> >
> > and a some interviews with the (re)inventors here:
> >
> > http://channel9.msdn.com/Showpost.aspx?postid=68302
> > http://channel9.msdn.com/Showpost.aspx?postid=141858
> >
> >
>



--
Nietzsche's first step is to accept what he knows. Atheism for him goes
without saying and is "contructive and
radical". Nietzsche's supreme vocation, so he says, is to provoke a kind of
crisis and a final decision about the
problem of atheism. The world continues on its course at random and there is
nothing final about it. Thus God
is useless, since He wants nothing in particular. If he wanted something --
and here we recognize the traditional
forumlation of the problem of evil -- He would have to assume responsiblity
for "a sum total of pain and inconsistency
which would debase the entire value of being born."
-- Albert Camus, L'Homme révolté

[-- Attachment #2: Type: text/html, Size: 2251 bytes --]

Re: [9fans] MS Research reinvents Inferno?

[Kenneth Long]

> On Mon, Dec 12, 2005 at 11:02:53PM -0800, Skip
> Tavakkolian wrote:
> [...]
> >
> http://channel9.msdn.com/Showpost.aspx?postid=68302
>          ^^^^^^^^
> 
> channel9 ? plan9?
> plan9? channel9?
> 

Theme music to the X-Files fades in...

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 

Re: [9fans] MS Research reinvents Inferno?

["Nils O. Selåsdal"]

Sergey Zhilkin wrote:
> Microsoft reinventing the wheels since 1980 :)
> Any sources of Sing ?
> Inferno IS better :) and also OpenSource :)
Yeah, but there's a key difference in that it isn't a product
of the infinite monkey theorem.

Re: [9fans] MS Research reinvents Inferno?

[Sergey Zhilkin]

Microsoft reinventing the wheels since 1980 :)
Any sources of Sing ?
Inferno IS better :) and also OpenSource :)

On 12/13/05, Skip Tavakkolian <9nut@9netics.com> wrote:
> I'm not sure if this is old news to you all. It's something
> that MS Research calls 'Singularity'.
>
> ftp://ftp.research.microsoft.com/pub/tr/TR-2005-135.pdf
>
> and a some interviews with the (re)inventors here:
>
> http://channel9.msdn.com/Showpost.aspx?postid=68302
> http://channel9.msdn.com/Showpost.aspx?postid=141858
>
>

Re: [9fans] MS Research reinvents Inferno?

[Oliver Bandel]

On Mon, Dec 12, 2005 at 11:02:53PM -0800, Skip Tavakkolian wrote:
[...]
> http://channel9.msdn.com/Showpost.aspx?postid=68302
         ^^^^^^^^

channel9 ? plan9?
plan9? channel9?


....


Ciao,
   Oliver

Re: [9fans] MS Research reinvents Inferno?

[Devon H. O'Dell]

2005/12/12, Skip Tavakkolian <9nut@9netics.com>:
> I'm not sure if this is old news to you all. It's something
> that MS Research calls 'Singularity'.
>
> ftp://ftp.research.microsoft.com/pub/tr/TR-2005-135.pdf
>
> and a some interviews with the (re)inventors here:
>
> http://channel9.msdn.com/Showpost.aspx?postid=68302
> http://channel9.msdn.com/Showpost.aspx?postid=141858

yeah, i read about 20 pages of that paper and tossed it aside. my
comment to the freebsd committer who pasted the url was ``plan 9.''

indeed, it does do a bit more with some things. but it's pure
research. and it's somewhat stupid anyway. not to dis plan 9, but this
sounds even less usable for most people.

--devon

p.s. please forgive my lack of a shift key, here. synergy decided to suck.

[9fans] MS Research reinvents Inferno?

[Skip Tavakkolian]

I'm not sure if this is old news to you all. It's something
that MS Research calls 'Singularity'.

ftp://ftp.research.microsoft.com/pub/tr/TR-2005-135.pdf

and a some interviews with the (re)inventors here:

http://channel9.msdn.com/Showpost.aspx?postid=68302
http://channel9.msdn.com/Showpost.aspx?postid=141858