Linux-Development-Sys Digest #779, Volume #6 Fri, 4 Jun 99 16:14:27 EDT
Contents:
Re: TAO: the ultimate OS (Vladimir Z. Nuri)
Re: TAO: the ultimate OS (Vladimir Z. Nuri)
Re: the ultimate OS (Vladimir Z. Nuri)
Re: TAO: the ultimate OS (Vladimir Z. Nuri)
Re: How to get all PID ? (Arun Sharma)
Re: TAO: the ultimate OS (Vladimir Z. Nuri)
Re: Might Linux SMP write memory out of order? ("Stefan Monnier "
<[EMAIL PROTECTED]>)
Tracing Program Counter (pradeep bollineni)
Re: Configuration Manager for Linux (Selious)
Re: A simple question...
(=?iso-8859-1?Q?Jos=E9=20Mar=EDa=20Fern=E1ndez=20Gonz=E1lez?=)
Linux daemons ("Baldbass")
Re: Exit status, $?, and term status (Justin Vallon)
Re: Linux daemons ("G. Sumner Hayes")
Re: TAO: the ultimate OS ("G. Sumner Hayes")
----------------------------------------------------------------------------
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
From: [EMAIL PROTECTED] (Vladimir Z. Nuri)
Subject: Re: TAO: the ultimate OS
Date: Fri, 4 Jun 1999 18:27:00 GMT
Christopher B. Browne ([EMAIL PROTECTED]) wrote:
: On Fri, 4 Jun 1999 03:20:49 GMT, Vladimir Z. Nuri <[EMAIL PROTECTED]>
: posted:
: > * object orientation is the best method we have for dealing with
: > complexity because it is the most complete form of abstraction
: Is it possible to get more "complete" explanation of what on earth
: "the most complete form of abstraction" represents? This sounds a
: whole lot like the usual rubbish that comes from people that want to
: be "ultimate" but that don't have a grasp of computing theory.
prior to OOP, there was a dichotomy between code and data.. OOP
gives you a kind of symmetry between code and data. it says the
basic unit of computing is an object, which contains both code
and data. earlier procedural programming used a form of data
hiding by saying a function or procedure was the basic unit
of programming, but then you have difficulties with stored state.
the object has a stored state. it's a reasonable /fundamental/
enlightened standard I am not challenging-- I'm saying it needs to move
outside of languages into the OS and file system itself.
: From what I can tell, the Lambda calculus and the concept of the
: Turing Machine are the "dual dueling masters" in the representation of
: the completeness of computing abstractions. Anything that can be
: computed may be expressed either as a Turing Machine or as an expression
: in the Lambda calculus. That's completeness...
ugh. familiar with both, but lets not introduce that into a
discussion of OSes ok? at least I have nothing I want to add.
the rest of your post is challenging comments someone else made.
--
~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^
"in theory, there's no difference [EMAIL PROTECTED]
between theory and practice, mad genius research lab
but in practice there is!" http://www8.pair.com/mnajtiv/
------------------------------
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
From: [EMAIL PROTECTED] (Vladimir Z. Nuri)
Subject: Re: TAO: the ultimate OS
Date: Fri, 4 Jun 1999 18:22:53 GMT
G. Sumner Hayes ([EMAIL PROTECTED]) wrote:
...
hi GSH, I seem to have popped out of your killfile already.
I'm not designing memes? excuse me? anyone who posts on
usenet is designing memes, imho. anyone who THINKS is
designing memes. which reminds me, "a great many people
get lost in thought because its such unfamiliar territory"
: But as I've been saying, you aren't designing memes. There's no
: evolution. You're repeating things that have been thoroughly
: discussed in the HCI community and not adding anything new.
where are the documents? I promise to read anything you
can some up with. HCI.. what is that? and btw .. discussion..
like others say, talk is cheap.. but an essay takes time
to write<g>
If
: you have no new ideas and no new code, you're just evangelizing --
: that's fine if you do it effectively, but judging from the reaction
: you're getting you need to change your approach if you really want
: to convince people.
what approach do you suggest, dear sir? do you believe it is
a worthwhile goal to come up with a design document that
represents a consensus, irrespective of code, prior to
code being written? such an activity seems so obviously
crucial to me, yet the linux community clearly holds it
in some contempt, feeling that code writing is the only
worthwhile/manly activity.. everyone else in the world
is a bonehead, hahahaha
: For example, do you understand why there was a boom in object
: databases (like you propose for the filesystem) about 5 years ago
: and why people have become much less interested in the idea today?
performance? I believe some of the problems with performance
can be remedied by makign the compiler an integral part of
the OS, as I wrote in the post. the idea is that all code
on the system can be recompiled on the fly depending on
how it is being used.
: Or the FS approach that Beos took? Do you understand the difference
: between the object models in C++, Python, and Smalltalk? How about
: signatures? Information-flow (leaving that term nebulous) interfaces
: as an alternative to object-oriented (idem) design?
there are many ways to implement objects. I'm familiar with the
basics (including a more recent language you don't mention, java)
notice that you are referring to languages, not OSes.
: Just screaming "make it all object oriented!" is worthless. It's
: not even clear what you mean by object oriented, but if you mean
: bundling data and code tightly together then there's a pretty
: strong consensus among HCI researchers that that is a terrible way
: to design a system if you're worried about the end user being able
: to do what he's trying to do without a fuss.
the point is to make the entire system based on objects, for
that to be the basic entity that is used by the end user in
code and operations. I don't know what you mean by "bundling
code and data together".. at least the idea I am envisioning
is a system in which the source code can be bundled with the
object, and you have a very powerful compiler that can
"take apart" & "put together" whenever you want, not merely
when you are creating the OS.
: There are conventions, CS departments, and an ACM SIG dedicated
: to the general topic that you're trying to tackle. Attacking
: the problem without understanding even the basic research of the
: last 20 years is foolish. I'm not about to give an HCI course
: on Usenet, but CHI and Carnegie Mellon's HCI institute are good
: starting places that link to other sites and a good number of papers.
believe we are talking about two different problems. I am not
trying to create a better object oriented paradigm for languages.
I am trying to leverage the already outstanding research into
objects to invade the OS itself, to find a unified system for
computation that increasingly dissolves the boundary between OS and
applictions.
BTW I don't believe you really read my paper.. nor that few
people here have.. it is not
merely saying, "make it all object oriented"
thanks for the urls
--
~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^
"in theory, there's no difference [EMAIL PROTECTED]
between theory and practice, mad genius research lab
but in practice there is!" http://www8.pair.com/mnajtiv/
------------------------------
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
From: [EMAIL PROTECTED] (Vladimir Z. Nuri)
Subject: Re: the ultimate OS
Date: Fri, 4 Jun 1999 18:07:43 GMT
definitely, there is a lot of frustration on this
topic, which my post & comments have brought out.
as for the traffic on the subject, its parly because I enjoy
fanning flames<g>.. also people tend to look more at
actual threads because they represent activity, so I like
to do my part.
"we all have a vision of what an OS should be".. true, but
how many people write it out? how many people attempt to harmonize
their goals before writing code? isn't this a reasonable
way to do it? a good design document is an incredibly powerful
focusing mechanism. someone said, "maybe all you have is a wish
list".. so what? a wish list that everyone agrees on is a
course of action.
building consensus before writing
code is a worthwhile activity imho... perhaps even crucial.
those who are into the "write a lot of code and call me when
you're done" school are free to ignore it, but I feel strongly
they are missing out.
I disagree that the open software movement would have
to "effectively sell its soul to microsoft" merely to adopt
"commercial software development practices".. such as
creating design documents, at least. (is that what
you are talking about?)
--
~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^
"in theory, there's no difference [EMAIL PROTECTED]
between theory and practice, mad genius research lab
but in practice there is!" http://www8.pair.com/mnajtiv/
------------------------------
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
From: [EMAIL PROTECTED] (Vladimir Z. Nuri)
Subject: Re: TAO: the ultimate OS
Date: Fri, 4 Jun 1999 18:40:26 GMT
G. Sumner Hayes ([EMAIL PROTECTED]) wrote:
: But there's a _start_ to project X, which gives people something to
: work with and a framework in which to evaluate the general approach.
: If you're into Internet anthropology, Eric Raymond has a paper
: discussing how open development works -- one key point he makes is
: that you have to start with enough code to get people excited before
: you start the "release early and often" cycle.
its a reasonable approach, has worked well,
but there is another way to do it. I've read the infamous
"cathedral vs. bazaar" model.. a rambling document if I ever
read one.. I was a bit shocked/annoyed with how meandering that thing
is and the fame it has received.. a design document IS INDEED A START
TO PROJECT X. and perhaps a far more sensible one than writing
code.
: I can remember exactly one project (KDE) that started with a post
: to Usenet proposing an idea with no code that actually ever got
: any serious development. And that was because the one who posted it
: got "show us the code" flames and went and did it.
*everything* starts with memes, not code. anyone who says otherwise
is .. er.. arguing against the obvious imho.
this whole "show us the code" thing reminds me of capitalism.
someone pitches idea for company. "I would like to do [x]
as a company". other people say, "yeah, so what?!?! go start
the company yourself". they don't contribute until they
see something happening. yes, often it boils down to a single
person who starts it. but it does make a lot of sense to get
together people who agree on the vision, don't you think?
what I think you are saying, is that you don't want to contribute
until you see MOMENTUM. for you, code is momentum. mere ideas
are not momentum. but to me, even mere ideas have momentum.
there are some people who are forward looking enough that they
will begin to contribute even when there is only an IDEA.
at the very earliest stage of a project, they are absolutely
crucial. I have run into a few such people who have sent
me mail.. I appreciate that.
everyone has a momentum threshold .. I respect that. some people
may hold off until a lot of momentum is built up in a project,
and then really engage, and perhaps so even more than the people
who started it.
let's just all agree we have different momentum threshholds,
ok? but arguing for one threshhold over another (i.e.
"ignore people with a low momentum threshhold, they do nothing
but yack without writing code") is unproductive imho.
yes, I am trying to attract people with low momentum threshholds
right now, and getting flamed by people with high ones.
there are some people so brilliant, like yang/filo, who
can commit radically to a vision without any momentum
whatsoever. they are clearly the most highly rewarded/
entrepreurial in the world. in fact the greatest entrepreneurs,
I would submit to you, have the lowest "momentum
threshholds". they start the project and commit to it
passionately/obsessively even before ANYTHING exists. they
create it out of thin air.
admittedly, crackpots/blowhards also have very low
"momentum threshholds" <g>
ok? my philosophizing for the day. hehehe
--
~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^
"in theory, there's no difference [EMAIL PROTECTED]
between theory and practice, mad genius research lab
but in practice there is!" http://www8.pair.com/mnajtiv/
------------------------------
Subject: Re: How to get all PID ?
From: Arun Sharma <[EMAIL PROTECTED]>
Date: Fri, 04 Jun 1999 05:31:55 GMT
man getpid
-Arun
------------------------------
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
From: [EMAIL PROTECTED] (Vladimir Z. Nuri)
Subject: Re: TAO: the ultimate OS
Date: Fri, 4 Jun 1999 18:29:48 GMT
Stefaan A Eeckels ([EMAIL PROTECTED]) wrote:
: Insofar as Linux is UNIX, it presents a pretty coherent vision
: (essentially, everyting is a file, small, cooperative programs
: around a text pipeline, etc). It's a tribute to the coherence
: of the UNIX vision that it is still relevant 30 years after it
: was conceived.
imho vision refers to the future, not the past. YMMV
: Why the obsession with Linux advancing further? It does what
: it does quite nicely. Those who want to use it, use it. Those
: who need new features, add them (or squeal).
an attempt to reach the next level..?
: Chris is right - the only way to kickstart a new, better OS
: is to write code. If it's good, then the history of Linux
: suggests that there will be lots of competent people who'll
: adopt it, and turn it into a fully fledged system.
again, obviously I don't debate the point that if you want
a working system, you must write code. I am saying, if you want
to write great code, perhaps there is even a step before
that.. which can be done in public.
"you can't run something that hasn't been written"...
wow........the mind reels.......NO KIDDING!!! methinks
I sense a collective blind spot
(arghghghg)
--
~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^
"in theory, there's no difference [EMAIL PROTECTED]
between theory and practice, mad genius research lab
but in practice there is!" http://www8.pair.com/mnajtiv/
------------------------------
From: "Stefan Monnier <[EMAIL PROTECTED]>"
<[EMAIL PROTECTED]>
Subject: Re: Might Linux SMP write memory out of order?
Date: 04 Jun 1999 14:20:52 -0400
>>>>> "David" == David Wragg <[EMAIL PROTECTED]> writes:
> x86 is just as (un)reasonable as Alpha, MIPS, PPC, Sparc, in this
> regard: If you access multiple shared variables without the protection
> of synchronisation primitives (mutexes etc.), you need memory barriers
> to avoid getting bitten.
But the guarantees are not all the same. MIPS has particularly strong
guarantees, while Alpha has them unusually weak.
For example (assuming a and b are both set to zero):
store a, 1 read b;
store b, 1 read a;
if `read b' returns 1, then `read a' is IIRC guaranteed by x86 to return 1 as
well.
> implied by machine code. The way to tell the compiler what you want is
> using volatile (or function calls). The way to tell the processor what
Hmmm... does ANSI C really enforce such a constraint on function calls ?
That would make it quite a bit harder for compilers to implement inlining
correctly.
Stefan
------------------------------
From: pradeep bollineni <[EMAIL PROTECTED]>
Subject: Tracing Program Counter
Date: Fri, 4 Jun 1999 00:37:40 -0500
Hi
Can someone tell me how to monitor a program so as to
get the following information:
The program counter value when a system call is made from the
monitored program.
Thanks
Pradeep
------------------------------
From: Selious <[EMAIL PROTECTED]>
Subject: Re: Configuration Manager for Linux
Date: Fri, 04 Jun 1999 21:16:35 +0200
>
> Both opinions have their strong and weak points and I wouldn't like you to think
> that I am advocating either of two.
Ehh, we don't do that !! It is impossible...
I mean, it would totally SPLIT up the current running UNIX/LINUX family, and the
new one.
Unless we all upgrade every machine on the year 2000 !!
No, I mean, a BIG advantage of linux is that applications like sendmail run under
UNIX !!
If linux becomes SO arrogant, it shall split up with UNIX.
But if we will, I will surely help !!
(my first idea to, but I don't see it happen).
Anyway, I rather have a event service in linux that notifies me of anything that
happens (for what I register) mixed in the kernel code.
Or fooling programs that /dev/hosts is a normal file on EXT2, but in fact a
/etc/hosts emulator that collects settings from this global database, and builds
the valid /etc/hosts on each query (or with a xx minute cache, or whatever
sceptisist can say about THIS. Please make AI and filter my news from secptisist
!!).
I have been thinking a little, but must admit my insights in the linux kernel are
limmited to the bootup and setup code, and I stil haven't reached the init process
startup jet !!
So to be honest, my project is more based on working ahead of my head by building
config file parsing and producing objects, researching code for 'official syntax',
and building a polling prototype that 'acts' as a event manager for the LMA.
Anyway, instead of killing one of us' brainstorms, I rather filosify about a 'Open
Object Handling Module' that can descibe itself by:
- monitored files and attributes
- monitored processes (is process running OK, is it zombie, gone, etc.)
- file size and checksum checks (is a file changed, like ls)
- volume free space
- directory size
- netstat
- w, who
- whatever we wanna be able to monitor and anticipate to remotely !!
All these events must be monitored by a event manager that simply tells registered
object handlers about a event taking place.
Than a security alarm can go off (ls changed !!), a central database can be updated
(a user get's a new telephone number), logfiles can be monitored (pretty close to
REAL events services), etc.
The problem with user logon's, process execution, etc. is that polling can miss
such a event, unless kernel code is adjusted to put these events (even if they took
one nanosecond) in a event queu, handled by the REAL event manager that can be set
to a certain priority to NOT slow done http operations and all while handling ALL
events on the linux machine, and not only the 'long lasting or logged' events
Ehh, RFC ???
------------------------------
From: =?iso-8859-1?Q?Jos=E9=20Mar=EDa=20Fern=E1ndez=20Gonz=E1lez?=
Subject: Re: A simple question...
Date: Fri, 04 Jun 1999 21:09:11 +0200
Hi again,
sorry, I don't know what it was happening with the DNS, but now I am able to
reach http://www.tomshardware.com.
--
Jos� Mar�a Fern�ndez Gonz�lez e-mail: [EMAIL PROTECTED]
Tlfn: (+34) 91 585 46 69 Fax: (+34) 91 585 45 06
Grupo de Dise�o de Proteinas Protein Design Group
Centro Nacional de Biotecnolog�a National Center of Biotechnology
C.P.: 28049 Zip Code: 28049
Campus Universidad Aut�noma. Cantoblanco, Madrid, Spain.
------------------------------
From: "Baldbass" <[EMAIL PROTECTED]>
Subject: Linux daemons
Date: Fri, 4 Jun 1999 14:09:16 -0500
Are there any places I can find information on writing a daemon for a Linux
newbie?
------------------------------
From: Justin Vallon <[EMAIL PROTECTED]>
Subject: Re: Exit status, $?, and term status
Date: 4 Jun 1999 00:20:20 -0400
John Nolan <[EMAIL PROTECTED]> writes:
> How can I tell whether $? contains the actual exit status
> of a program or its termination signal and coredump flag?
>
> For example:
>
> [A] If I run "ls -l zzzz", where the file zzzz does
> not exist, then $? contains "2". This is the exit
> status of ls.
>
> [B] However, if I run a small compiled C program like this
>
> int main() {
>
> int i;
> i = 1/0; /* Floating point exception, signal 8 */
> }
>
> ....then $? contains "136". 136 is not actually
> the exit status, it is the signal ANDed with
> the coredump flag. (128 && 8, or in hex: 0x80 && 0x08).
>
> Or, to rewrite the term status in binary:
>
> [A] term status = 0000 0010 0000 0000
> [B] term status = 0000 0000 1000 1000
As you probably know, these values are portably checked (in C) via the
wait-macros: WIFEXITED(status), WEXITSTATUS(status),
WIFSIGNALED(status), WTERMSIG(status). See man wait.
The value that appears in $? is put there by the shell. Check bash,
tcsh, sh, etc, for the rules. bash indirectly describes what appears
in $? via its "Exit Status" section:
EXIT STATUS
For the purposes of the shell, a command which exits with
a zero exit status has succeeded. An exit status of zero
indicates success. A non-zero exit status indicates fail-
ure. When a command terminates on a fatal signal, bash
uses the value of 128+signal as the exit status.
If a command is not found, the child process created to
execute it returns a status of 127. If a command is found
but is not executable, the return status is 126.
Bash itself returns the exit status of the last command
executed, unless a syntax error occurs, in which case it
exits with a non-zero value. See also the exit builtin
command below.
--
-Justin
[EMAIL PROTECTED]
------------------------------
From: "G. Sumner Hayes" <[EMAIL PROTECTED]>
Subject: Re: Linux daemons
Date: Fri, 04 Jun 1999 15:47:40 -0400
Baldbass wrote:
> Are there any places I can find information on writing a daemon for
> a Linux newbie?
Read the comp.unix.programmer FAQ. It discusses the general concepts
behind making your program a daemon.
http://www.erlenstar.demon.co.uk/unix/faq_toc.html
--Sumner
------------------------------
From: "G. Sumner Hayes" <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: TAO: the ultimate OS
Date: Fri, 04 Jun 1999 03:09:44 -0400
"Christopher B. Browne" wrote:
> From what I can tell, the Lambda calculus and the concept of the
> Turing Machine are the "dual dueling masters" in the representation of
> the completeness of computing abstractions. Anything that can be
> computed may be expressed either as a Turing Machine or as an expression
> in the Lambda calculus. That's completeness...
Nit: "Dual dueling masters" seems to imply that these are the only
two. There are plenty of others -- URMs are a common one used in
introductory logic & computation theory courses. It's a lot more
fun to program for an URM than a Turing Machine, so after proving
the equivalence of the two many classes just stick with URMs and
the Lambda calculus.
--Sumner
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.development.system) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Development-System Digest
******************************
