Ada-inspired OS/Language

Ada-inspired OS/Language

[Ryan]

An article on Slashdot reads
(http://developers.slashdot.org/article.pl?sid=02/08/18/2053259)

"Brand Huntsman (the creator of the Bochs Front-End, among other obscure 
things) has been developing an integrated language/operating system for 
the past few years now. The Operating System is called BRiX, and it uses 
a language called Crush, which is woven tightly into the core of the OS. 
On his project web page (http://brix-os.sf.net) he has posted the source 
code (http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/brix-os/crush) to 
his preliminary compiler, which runs in Linux and outputs optimized 
assembly from Crush source code. The Crush language itself is heavily 
influenced by Forth, LISP, and Ada, and provides strong typing and 
extensive namespace security."

An interesting quote from the FAQ on his site is

"BRiX, unlike other modern operating systems, does not use hardware to 
isolate and protect applications from each other. Instead, it uses a 
single address space and relies on a safe-language to generate code that 
will not access memory for which it does not own. This language also 
handles many checks at compile-time that would be performed at run-time 
in other operating systems."

What effect might this have on the development of AdaOS?  Competition? 
Inspiration?  It seems to have several of the same goals in mind.

Ryan

Re: Ada-inspired OS/Language

[Jim Rogers]

Ryan wrote:

> An article on Slashdot reads
> (http://developers.slashdot.org/article.pl?sid=02/08/18/2053259)
> 
> "Brand Huntsman (the creator of the Bochs Front-End, among other obscure 
> things) has been developing an integrated language/operating system for 
> the past few years now. The Operating System is called BRiX, and it uses 
> a language called Crush, which is woven tightly into the core of the OS. 
> On his project web page (http://brix-os.sf.net) he has posted the source 
> code (http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/brix-os/crush) to 
> his preliminary compiler, which runs in Linux and outputs optimized 
> assembly from Crush source code. The Crush language itself is heavily 
> influenced by Forth, LISP, and Ada, and provides strong typing and 
> extensive namespace security."
> 
> An interesting quote from the FAQ on his site is
> 
> "BRiX, unlike other modern operating systems, does not use hardware to 
> isolate and protect applications from each other. Instead, it uses a 
> single address space and relies on a safe-language to generate code that 
> will not access memory for which it does not own. This language also 
> handles many checks at compile-time that would be performed at run-time 
> in other operating systems."
> 
> What effect might this have on the development of AdaOS?  Competition? 
> Inspiration?  It seems to have several of the same goals in mind.
> 
> Ryan
> 

If you look at Crush you will see it more closely resembles Lisp than
Ada.

I suspect this effort may provide some interesting ideas. It also will
provide an example of an interesting OS that is largely ignored by the
world.

Jim Rogers

Re: Ada-inspired OS/Language

[Darren New]

Ryan wrote:
> "BRiX, unlike other modern operating systems, does not use hardware to
> isolate and protect applications from each other. 

This is how the Burroughs B-series computers worked.

Of course, one problem is that it completely eliminates the ability to use
any legacy code in any legacy language. You have a chicken-and-egg problem,
except you have neither chickens nor eggs to start with.

-- 
Darren New 
San Diego, CA, USA (PST). Cryptokeys on demand.
   ** http://images.fbrtech.com/dnew/ **

Try our EbolaBurgers...
      So tender they melt in your mouth.

Re: Ada-inspired OS/Language

[Warren W. Gay VE3WWG]

Ryan wrote:
> An article on Slashdot reads
> (http://developers.slashdot.org/article.pl?sid=02/08/18/2053259)
...
> "BRiX, unlike other modern operating systems, does not use hardware to 
> isolate and protect applications from each other. Instead, it uses a 
> single address space and relies on a safe-language to generate code that 
> will not access memory for which it does not own. 

Here we go again.. another DOS-like, memory wide open to corruption
concept. PC's eventually, with the help of 286 and later got to the
point where they could protect processes. Now we've regressed with
the palms etc., back to this no-protection model. No matter how good
the language is, there will always be occaisions where one process
steps on another unless it is prevented from doing so in hardware.

This is OK for research. Not good for running systems that must
run as firewalls, or central hosts to university students!

This model has been tried before BTW, with hardware protection
(each process just has system wide unique addresses for memory
objects). IIRC, one such system was called Opal.

This language also
> handles many checks at compile-time that would be performed at run-time 
> in other operating systems."
> 
> What effect might this have on the development of AdaOS?  Competition? 
> Inspiration?  It seems to have several of the same goals in mind.
> 
> Ryan

IMHO, I don't think it amounts to anything more than a research effort.
I can't see the world going to a less secure model (ie. no hardware
memory protection), after seeing all the fighting with the security
problems we already have in a more secure environment.

I think there is a vast amount of work that could be done in the
interest of a secure host O/S, that is [mostly?] immune to DOS
attacks that can be launched by guest university accounts, for
example.

-- 
Warren W. Gay VE3WWG
http://home.cogeco.ca/~ve3wwg

Re: Ada-inspired OS/Language

[Nick Roberts]

On Tue, 20 Aug 2002 13:57:24 -0400, "Warren W. Gay VE3WWG"
<ve3wwg@cogeco.ca> strongly typed:

>Ryan wrote:
>> An article on Slashdot reads
>> (http://developers.slashdot.org/article.pl?sid=02/08/18/2053259)
>...
>> "BRiX, unlike other modern operating systems, does not use hardware to 
>> isolate and protect applications from each other. Instead, it uses a 
>> single address space and relies on a safe-language to generate code that 
>> will not access memory for which it does not own. 
>
>Here we go again.. another DOS-like, memory wide open to corruption
>concept. PC's eventually, with the help of 286 and later got to the
>point where they could protect processes. Now we've regressed with
>the palms etc., back to this no-protection model. No matter how good
>the language is, there will always be occaisions where one process
>steps on another unless it is prevented from doing so in hardware.
>
>This is OK for research. Not good for running systems that must
>run as firewalls, or central hosts to university students!

This is one reason why BRiX (or Oberon) will not have a tremendous effect
on the design of AdaOS.

The other main reason is that this type of design appears to preclude the
possibility of the direct execution of programs imported in machine code
form. I find this an unacceptable restriction, and I think too many users
would as well.

By the way, there is a simple overview of AdaOS now available at

http://www.adaos.net

with a pretty diagram if anybody is interested.

-- 
Nick Roberts

Re: Ada-inspired OS/Language

[Marin D. Condic]

It looks interesting. Does this mean that development on AdaOS is ramping
up? Here's one thing I think is an excellent hook: The fact that you are
aiming to make it a distributed system. Figure that there are any number of
geeks out there with two or more old PCs sitting around the house who might
find it really fun to put an OS on them that turned them into a cluster.
That's something kind of unique. Perhaps the more it aims at distribution
and load balancing the more of a niche it will find in industry. If it makes
scaling up a system easy by just plugging in more inexpensive computers, it
could really catch on. Especially to the extent that it can exploit Ada
tasking as a distribution mechanism - "Sure, you can program it in C/C++ but
it'll be easier to exploit parallelism and get better performance if you
just do it in Ada..."

BTW: What theme is driving the naming scheme?

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com

Enabling Digital.
Our Vision is to be the biggest supplier worldwide of digital gateway
technology.
www.pacemicro.com

"Nick Roberts" <nickroberts@blueyonder.co.uk> wrote in message
news:3d6e9cb6.75108980@news.cis.dfn.de...
> On Tue, 20 Aug 2002 13:57:24 -0400, "Warren W. Gay VE3WWG"
> <ve3wwg@cogeco.ca> strongly typed:
>
> >Ryan wrote:
> >> An article on Slashdot reads
> >> (http://developers.slashdot.org/article.pl?sid=02/08/18/2053259)
> >...
> >> "BRiX, unlike other modern operating systems, does not use hardware to
> >> isolate and protect applications from each other. Instead, it uses a
> >> single address space and relies on a safe-language to generate code
that
> >> will not access memory for which it does not own.
> >
> >Here we go again.. another DOS-like, memory wide open to corruption
> >concept. PC's eventually, with the help of 286 and later got to the
> >point where they could protect processes. Now we've regressed with
> >the palms etc., back to this no-protection model. No matter how good
> >the language is, there will always be occaisions where one process
> >steps on another unless it is prevented from doing so in hardware.
> >
> >This is OK for research. Not good for running systems that must
> >run as firewalls, or central hosts to university students!
>
> This is one reason why BRiX (or Oberon) will not have a tremendous effect
> on the design of AdaOS.
>
> The other main reason is that this type of design appears to preclude the
> possibility of the direct execution of programs imported in machine code
> form. I find this an unacceptable restriction, and I think too many users
> would as well.
>
> By the way, there is a simple overview of AdaOS now available at
>
> http://www.adaos.net
>
> with a pretty diagram if anybody is interested.
>
> --
> Nick Roberts

Re: Ada-inspired OS/Language

[Nick Roberts]

On Fri, 30 Aug 2002 09:33:11 -0400, "Marin D. Condic" <not.valid@acm.org>
strongly typed:

>It looks interesting. Does this mean that development on AdaOS is ramping
>up? 

I'm not sure about "ramping up"! Maybe stepping up a gear. I can only hope
that as people see the vast technical superiority (and I'm not being
ironic, for once) of AdaOS, they will be attracted. But that will surely be
a slow process for the next few years.

>Here's one thing I think is an excellent hook: The fact that you are
>aiming to make it a distributed system. ... That's something kind of unique. 

Yes, I think it is. I am not very familiar with Andrew, but my impression
is that in fact no-one has produced a truly distributed OS yet (that can
shuffle processes between workstations arbitrarily, without any major
restrictions).

>Perhaps the more it aims at distribution and load balancing the more of a 
>niche it will find in industry. If it makes scaling up a system easy by 
>just plugging in more inexpensive computers, it could really catch on. 

I believe AdaOS really will make it easy to just add on computers. But this
is not the only potential hook. It seems AdaOS may be the first general OS
to offer proper security facilities (TCSEC B1 or B2). Do any others offer
full dual labelling?

>Especially to the extent that it can exploit Ada tasking as a distribution 
>mechanism ...

I'm glad you have seen that point. I think it is a very significant one.
There is currently a hot market in Internet servers, for e-business and
many other purposes. These still all work on software that is very
primitive (e.g. vulnerable to buffer overrun vulnerabilities), and I see
this as one potential ouvre for AdaOS.

>BTW: What theme is driving the naming scheme?

Some of the names are old, and have simply stuck. For example, the program
named "Avarus" (Latin for "miser") was named long before it became the
object system manager, and was instead responsible for storage compaction.

Some are new, and I have endeavoured to choose names that are original but
which have some relevance. "Bachar" is the Gaelic for "acorn", which seemed
appropriate for a microkernel. "Belenus" was a Celtic god of light and fire
(and he was also associated with the forge and metalworking), and so seemed
quite appropriate for a windowing program. "Carrot" is the name of a
character, from the Discworld series of books by Terry Pratchett, who is
phenomenal at remembering people's names (quite unlike myself ;-).
Suggestions for program names are always welcome!

-- 
Nick Roberts

Re: Ada-inspired OS/Language

[Marin David Condic]

Well, there you have it. Since lots of servers depend on tasks/threads for the
mechanism of communicating with clients, being able to add processors and
increase performance is a serious advantage. You can get there now I'm sure, but
at what level of difficulty? Having it happen automagically is something people
will like.


MDC
--
=============================================================
Marin David Condic - Pace Micro - http://www.pacemicro.com/
Send Replies To: m c o n d i c @ a c m . o r g
Visit my web site at:  http://www.mcondic.com/

    "I'd trade it all for just a little more"
        --  Charles Montgomery Burns, [4F10]
=============================================================



Nick Roberts wrote:

> >Especially to the extent that it can exploit Ada tasking as a distribution
> >mechanism ...
>
> I'm glad you have seen that point. I think it is a very significant one.
> There is currently a hot market in Internet servers, for e-business and
> many other purposes. These still all work on software that is very
> primitive (e.g. vulnerable to buffer overrun vulnerabilities), and I see
> this as one potential ouvre for AdaOS.
>
>

--
=============================================================
Marin David Condic - Pace Micro - http://www.pacemicro.com/
Send Replies To: m c o n d i c @ a c m . o r g
Visit my web site at:  http://www.mcondic.com/

    "I'd trade it all for just a little more"
        --  Charles Montgomery Burns, [4F10]
=============================================================

Re: Ada-inspired OS/Language

[Jacob Sparre Andersen]

Nick Roberts wrote:

> Yes, I think it is. I am not very familiar with Andrew, but my impression
> is that in fact no-one has produced a truly distributed OS yet (that can
> shuffle processes between workstations arbitrarily, without any major
> restrictions).

IIRC Tao (or is it Taos?) can do this. And even in a 
partially architecture independent way. But I haven't played 
with it for a very long time.

Jacob
-- 
Do you have a job for a physicist with experience in:

  * Complex systems (mostly turbulence experiments)
  * Geoscience (a bit of oceanography and geology)
  * Programming (mostly in Ada and Pascal)

Re: Ada-inspired OS/Language

[Darren New]

Nick Roberts wrote:
> Yes, I think it is. I am not very familiar with Andrew, but my impression
> is that in fact no-one has produced a truly distributed OS yet (that can
> shuffle processes between workstations arbitrarily, without any major
> restrictions).

I believe Ameoba (by Tanenbaum) will do this without much trouble. Of
course, you have to move the process to a machine with the same CPU and
enough memory, but that seems reasonable. It's a capability-based OS, and it
assumes (for example) that the place you compute and the place you hand
disks off of are two separate machines. (I think it even assumes your
directories live on a different machine than the actual data files.) Moving
processes between computers would be a user-level operation in this system.
(Freeze the process, find a new computer, allocate new memory, copy old
memory to new memory, start process in new memory, abort old process.)

Google for it. The papers describing the architecture are (were) online, if
not the full source code.

-- 
Darren New 
San Diego, CA, USA (PST). Cryptokeys on demand.
   ** http://images.fbrtech.com/dnew/ **

Try our EbolaBurgers...
      So tender they melt in your mouth.

Re: Ada-inspired OS/Language

[Nick Roberts]

On Wed, 04 Sep 2002 17:24:32 GMT, Darren New <dnew@san.rr.com> strongly
typed:

>Nick Roberts wrote:
>> Yes, I think it is. I am not very familiar with Andrew, but my impression
>> is that in fact no-one has produced a truly distributed OS yet (that can
>> shuffle processes between workstations arbitrarily, without any major
>> restrictions).
>
>I believe Ameoba (by Tanenbaum) will do this without much trouble. Of
>course, you have to move the process to a machine with the same CPU and
>enough memory, but that seems reasonable. It's a capability-based OS, and it
>assumes (for example) that the place you compute and the place you hand
>disks off of are two separate machines. (I think it even assumes your
>directories live on a different machine than the actual data files.) Moving
>processes between computers would be a user-level operation in this system.
>(Freeze the process, find a new computer, allocate new memory, copy old
>memory to new memory, start process in new memory, abort old process.)

This sounds very similar to how AdaOS will do it.

>Google for it. The papers describing the architecture are (were) online, if
>not the full source code.

I'm sure I tried this (and got relatively little information), but I will
try again! ... I tried and got loads of stuff. Great! Much reading matter.

-- 
Nick Roberts

Re: Ada-inspired OS/Language

[Georg Bauhaus]

Nick Roberts <nickroberts@blueyonder.co.uk> wrote:
:>Google for it. The papers describing the architecture are (were) online, if
:>not the full source code.
: 
: I'm sure I tried this (and got relatively little information), but I will
: try again! ... I tried and got loads of stuff. Great! Much reading matter.
 
http://www.cs.vu.nl/~ast/cs/dist.html

-- georg

Re: Ada-inspired OS/Language

[Rod Haper]

Nick Roberts wrote:
> On Wed, 04 Sep 2002 17:24:32 GMT, Darren New <dnew@san.rr.com> strongly
> typed:
> 
> 
>>Nick Roberts wrote:
>>
>>>Yes, I think it is. I am not very familiar with Andrew, but my impression
>>>is that in fact no-one has produced a truly distributed OS yet (that can
>>>shuffle processes between workstations arbitrarily, without any major
>>>restrictions).
>>
>>I believe Ameoba (by Tanenbaum) will do this without much trouble. Of
>>course, you have to move the process to a machine with the same CPU and
>>enough memory, but that seems reasonable. It's a capability-based OS, and it
>>assumes (for example) that the place you compute and the place you hand
>>disks off of are two separate machines. (I think it even assumes your
>>directories live on a different machine than the actual data files.) Moving
>>processes between computers would be a user-level operation in this system.
>>(Freeze the process, find a new computer, allocate new memory, copy old
>>memory to new memory, start process in new memory, abort old process.)
> 
> 
> This sounds very similar to how AdaOS will do it.
> 
> 
>>Google for it. The papers describing the architecture are (were) online, if
>>not the full source code.
> 
> 
> I'm sure I tried this (and got relatively little information), but I will
> try again! ... I tried and got loads of stuff. Great! Much reading matter.
> 


If you haven't already, you might also want to have a look at Inferno 
for ideas about a distributed OS architecture, applications, security, 
programming language support, etc.

http://www.vitanuova.com/inferno/index.html
http://www.vitanuova.com/inferno/concepts.html
http://www.vitanuova.com/inferno/papers.html
http://www.vitanuova.com/inferno/papers/bltj.html


-- 
Rod

Re: Ada-inspired OS/Language

[Richard Riehle]

Nick Roberts wrote:

> By the way, there is a simple overview of AdaOS now available at
>
> http://www.adaos.net

The architecture looks pretty good.   Hope you can bring
it to completion.

One comment.  I am concerned about calling it AdaOS.  At
present, anything with the name Ada in it is likely to be
ignored by the larger audience you would hope to gain.

The published product would do well to stand on its own,
support languages besides Ada, and have a name that
makes Ada less of an explicit issue.   In time, people
concerned with the technology will realize that it is
written in Ada, while those who simply want to get
some work done will use it without any knowledge
of its underlying implementation.

Meanwhile, if this project gets to the point where we can
see it becoming a reality, some of us ought to get busy
writing applications for it.   The only way it will become
popular is if real people use it.   Those real people will
need spreadsheets, word processors, etc.

Richard Riehle

Re: Ada-inspired OS/Language

[Nick Roberts]

On Fri, 06 Sep 2002 10:31:26 -0700, Richard Riehle <richard@adaworks.com>
strongly typed:

>Nick Roberts wrote:
>
>> By the way, there is a simple overview of AdaOS now available at
>>
>> http://www.adaos.net
>
>The architecture looks pretty good.   Hope you can bring
>it to completion.

Thanks, and I shall try.

>One comment.  I am concerned about calling it AdaOS.  At
>present, anything with the name Ada in it is likely to be
>ignored by the larger audience you would hope to gain.
>
>The published product would do well to stand on its own,
>support languages besides Ada, and have a name that
>makes Ada less of an explicit issue.   In time, people
>concerned with the technology will realize that it is
>written in Ada, while those who simply want to get
>some work done will use it without any knowledge
>of its underlying implementation.

I rather agree with you. AdaOS was originally named 'Alguinntar', which was
perhaps not the most inspiring name! Shortly after the project was formed
(in 1999), the members (then about 15) had a vote on the name, and the
overwhelming majority voted for 'AdaOS'.

In fact, there are some advantages. Occasionally someone will ask "Is that
AdaOS as in the Ada language I've heard of somewhere?," and this naturally
gives me the perfect ouvre for my (now polished ;-) 'sales' spiel about
Ada.

In fact, nearer the time, what I might do is to invent various 'marketing'
names for different releases of AdaOS. How many names does Linux have now?

I intend to implement a judicious selection of other languages on AdaOS.
For example, SQL3 or OQL for the database engine, and a C compiler. Maybe a
JVM.

>Meanwhile, if this project gets to the point where we can
>see it becoming a reality, some of us ought to get busy
>writing applications for it.   The only way it will become
>popular is if real people use it.   Those real people will
>need spreadsheets, word processors, etc.

Absolutely. AdaOS will be an object-oriented OS. This will provide a
framework for componentware, which I am sure is the key to getting
sophisticated application software out there without going through the
classic heavy programming cycle first.

I am not certain, but if AdaOS is based on CORBA 3, the opportunity will be
there for the direct import of COM components. Now we all know how reliable
typical COM components tend to be ;-) but this could be handy.

-- 
Nick Roberts

Re: Ada-inspired OS/Language

[Marin David Condic]

I can see Richard's point about "AdaOS" possibly having connotations that might
be off-putting to some people. However, far more important than the name is
that there be some usable, working piece of it in existence. Concentrate on
getting something that can boot up on an average PC and do *something*
semi-useful and you will have accomplished a lot. Even if all it will do is
fire up a command line and load some simple programs, you will have
accomplished a lot and provided something on which to build. Getting too
ambitious too soon is just going to kill it. Stick to some subset of stuff that
will enable the hobbyist to use easily available tools (A PC and the Gnat
compiler are a good start...) to a) develop pieces of the OS and b) develop
apps for the OS and you will thus attract the developers to get the effort
moved along. Changing the name later is no problem. Rename it "NickOS" if it
seems to sell better that way - but if there isn't anything to sell, the name
won't matter much.

I'd be impressed if you had assembled a development kit, instructions and some
minimal code that would serve as the starting point for an OS. Something to
compile a bootstrap program and get it  on a floppy that when power-cycling the
PC would start up & load some other program. You need at least that much to
start any serious OS development and getting that far would constitute "A Good
Start". Getting the tools together and a bootstrap going isn't going to impact
the overall design of the OS, so you don't have to have that thoroughly
designed in order to get started.

MDC
--
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jast.mil/

Send Replies To: m c o n d i c @ a c m . o r g

    "I'd trade it all for just a little more"
        --  Charles Montgomery Burns, [4F10]
======================================================================



Nick Roberts wrote:

>
> I rather agree with you. AdaOS was originally named 'Alguinntar', which was
> perhaps not the most inspiring name! Shortly after the project was formed
> (in 1999), the members (then about 15) had a vote on the name, and the
> overwhelming majority voted for 'AdaOS'.
>
> I

Re: Ada-inspired OS/Language

[Nick Roberts]

On Sat, 07 Sep 2002 09:43:33 -0400, Marin David Condic
<mcondic.auntie.spam@acm.org> strongly typed:

>I can see Richard's point about "AdaOS" possibly having connotations that might
>be off-putting to some people. However, far more important than the name is
>that there be some usable, working piece of it in existence. [etc.]

Yes I agree with this idea too. I have started a mini-project on
SourceForge just for the microkernel, called "Bachar". I shall try to get
this component done ASAP, and it will stand alone and could be usuable by a
lot of people for a variety of projects.

I am trying to square the circle of releasing a usable item as early as
possible, but at the same time ensuring that it will be of the highest
quality.

-- 
Nick Roberts

RE: Ada-inspired OS/Language

[Robert C. Leif]

If Nick Roberts et al. wish to generate some real interest and the
ability to raise money, call it XMLOS. The GUI would be XML and the
insides pure Ada. This is essentially what I suggested in my paper in
the last Ada Letters.
Bob Leif

-----Original Message-----
From: comp.lang.ada-admin@ada.eu.org
[mailto:comp.lang.ada-admin@ada.eu.org] On Behalf Of Marin David Condic
Sent: Saturday, September 07, 2002 6:44 AM
To: comp.lang.ada@ada.eu.org
Subject: Re: Ada-inspired OS/Language

I can see Richard's point about "AdaOS" possibly having connotations
that might
be off-putting to some people. However, far more important than the name
is
that there be some usable, working piece of it in existence. Concentrate
on
getting something that can boot up on an average PC and do *something*
semi-useful and you will have accomplished a lot. Even if all it will do
is
fire up a command line and load some simple programs, you will have
accomplished a lot and provided something on which to build. Getting too
ambitious too soon is just going to kill it. Stick to some subset of
stuff that
will enable the hobbyist to use easily available tools (A PC and the
Gnat
compiler are a good start...) to a) develop pieces of the OS and b)
develop
apps for the OS and you will thus attract the developers to get the
effort
moved along. Changing the name later is no problem. Rename it "NickOS"
if it
seems to sell better that way - but if there isn't anything to sell, the
name
won't matter much.

I'd be impressed if you had assembled a development kit, instructions
and some
minimal code that would serve as the starting point for an OS. Something
to
compile a bootstrap program and get it  on a floppy that when
power-cycling the
PC would start up & load some other program. You need at least that much
to
start any serious OS development and getting that far would constitute
"A Good
Start". Getting the tools together and a bootstrap going isn't going to
impact
the overall design of the OS, so you don't have to have that thoroughly
designed in order to get started.

MDC
--
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jast.mil/

Send Replies To: m c o n d i c @ a c m . o r g

    "I'd trade it all for just a little more"
        --  Charles Montgomery Burns, [4F10]
======================================================================



Nick Roberts wrote:

>
> I rather agree with you. AdaOS was originally named 'Alguinntar',
which was
> perhaps not the most inspiring name! Shortly after the project was
formed
> (in 1999), the members (then about 15) had a vote on the name, and the
> overwhelming majority voted for 'AdaOS'.
>
> I

Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Preben Randhol]

On Sun, 8 Sep 2002 15:14:13 -0700, Robert C. Leif wrote:
> If Nick Roberts et al. wish to generate some real interest and the
> ability to raise money, call it XMLOS. The GUI would be XML and the
> insides pure Ada. This is essentially what I suggested in my paper in
> the last Ada Letters.

What is the advantage of having the GUI as XML? I don't get it. Wouldn't
you need a more dynamic language than Ada then?

Preben

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Larry Kilgallen]

In article <slrnanots5.2b2.randhol+news@kiuk0156.chembio.ntnu.no>, Preben Randhol <randhol+news@pvv.org> writes:
> On Sun, 8 Sep 2002 15:14:13 -0700, Robert C. Leif wrote:
>> If Nick Roberts et al. wish to generate some real interest and the
>> ability to raise money, call it XMLOS. The GUI would be XML and the
>> insides pure Ada. This is essentially what I suggested in my paper in
>> the last Ada Letters.
> 
> What is the advantage of having the GUI as XML?

I think the point is that XML is trendy.

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Marin David Condic]

Trendy is nice, but there could be a technical advantage as well. Suppose you had an underlying Ada OS that had as
one of its parts a graphics engine that interpreted XML-based markup to control the screen. The OS and any apps
simply send properly formed markup to the engine to get things rendered and receive back XML markup for user
supplied data. Assuming the scheme is well thought out & executed, you get some nice advantages. Change the engine
and you can change the look&feel of all the apps - easier to create distinction between different OS releases. If
you use an existing XML standard the apps become instantly portable. If the interface to the engine is properly
thought out, it should be no big deal to make an app work across a network. Another OS could run the apps simply by
providing a compatible engine. From a low level an app can simply send/receive&parse XML, but a skin could be built
on top of it to provide a more traditional calling interface for a windowing scheme. Build a different skin - get a
different API with possibly different features.

The advantage is that you can disconnect the apps/OS from the graphics display through an intermediary language,
much like what web browsers do for the Internet with html.

MDC
--
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jast.mil/

Send Replies To: m c o n d i c @ a c m . o r g

    "I'd trade it all for just a little more"
        --  Charles Montgomery Burns, [4F10]
======================================================================



Larry Kilgallen wrote:

>
>
> I think the point is that XML is trendy.

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Preben Randhol]

On Mon, 09 Sep 2002 10:02:11 -0400, Marin David Condic wrote:

> Trendy is nice, but there could be a technical advantage as well.
> Suppose you had an underlying Ada OS that had as one of its parts a
> graphics engine that interpreted XML-based markup to control the
> screen. The OS and any apps simply send properly formed markup to the
                                          ^^^^^^^^
                                          Yes which to me sounds like a
                                          place were one can run into
                                          problems. Wouldn't this also
                                          increase overheads like
                                          run-time checks/validation and
                                          increased memory usage?

> engine to get things rendered and receive back XML markup for user
> supplied data. 

> The advantage is that you can disconnect the apps/OS from the graphics
> display through an intermediary language, much like what web browsers
> do for the Internet with html.

But if you have an application, say a program to store cooking recipies,
how can you attach the callbacks to the backend? I mean you cannot
simply just put in an extra dialog in the XML without coding support for
it in the backend?

Preben who is waiting for XML viruses

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Ryan Tarpine]

Marin David Condic wrote:
> Trendy is nice, but there could be a technical advantage as well. Suppose you had an underlying Ada OS that had as
> one of its parts a graphics engine that interpreted XML-based markup to control the screen. The OS and any apps

Sounds similar to libglade for the GTK widget set on *nix/*BSD:
   http://freshmeat.net/projects/libglade/?topic_id=58

"Libglade is a small library that allows a program to load its user 
interface from an XML description at runtime"

> simply send properly formed markup to the engine to get things rendered and receive back XML markup for user
> supplied data. Assuming the scheme is well thought out & executed, you get some nice advantages. Change the engine
> and you can change the look&feel of all the apps - easier to create distinction between different OS releases. If
> you use an existing XML standard the apps become instantly portable. If the interface to the engine is properly

"The XML file format is that of the user interface builder GLADE..."

I don't know anything about GLADE, but that could be considered as an 
existing "standard".

> thought out, it should be no big deal to make an app work across a network. Another OS could run the apps simply by
> providing a compatible engine. From a low level an app can simply send/receive&parse XML, but a skin could be built
> on top of it to provide a more traditional calling interface for a windowing scheme. Build a different skin - get a
> different API with possibly different features.

"Libglade also provides a simple interface for connecting handlers to 
the various signals in the interface"

> 
> The advantage is that you can disconnect the apps/OS from the graphics display through an intermediary language,
> much like what web browsers do for the Internet with html.
> 
> MDC

I think XML is a good solution to the problem of UI.  There are many 
places XML is being used that is should not be (see 
http://ask.slashdot.org/article.pl?sid=02/07/15/212226&mode=thread&tid=99 
for it being used as a network protocol: incredible amounts of bandwidth 
-- often more precious than CPU cycles -- wasted), but I don't think 
this is one.

Ryan

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Warren W. Gay VE3WWG]

Marin David Condic wrote:
> Trendy is nice, but there could be a technical advantage as well. 

"Trendy" can work against you too. When everyone gets
sick of hearing about XML (which is just a data format),
then this'll make the O/S smell.

I personally think that the name AdaOS raises a level
of curiosity (maybe its just me). Even hard core C/C++
people might be inclined to say to themselves "let's
see what those Ada folks are up to". To Ada enthusiasts
of course, this acts as a beacon ;-)

-- 
Warren W. Gay VE3WWG
http://home.cogeco.ca/~ve3wwg

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Georg Bauhaus]

Warren W. Gay VE3WWG <ve3wwg@cogeco.ca> wrote:
: XML (which is just a data format),

No. Actually this characterization is quite far off.
The fact that many people think XML is a data format
doesn't make it one. You can use the "XML declarative
toolset (with types, and with support for static and
runtime type checking)" to define data formats, via a
grammar specification. L stands for Language, and there
is more to it than bracketted words looking like 
"syntax clutter". SCNR.

Would anyone want to use C-OS?

-- Georg

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Warren W. Gay VE3WWG]

Georg Bauhaus wrote:
> Warren W. Gay VE3WWG <ve3wwg@cogeco.ca> wrote:
> : XML (which is just a data format),
> 
> No. Actually this characterization is quite far off.
> The fact that many people think XML is a data format
> doesn't make it one. You can use the "XML declarative
> toolset (with types, and with support for static and
> runtime type checking)" to define data formats, via a
> grammar specification. L stands for Language, and there
> is more to it than bracketted words looking like 
> "syntax clutter". SCNR.
> 
> Would anyone want to use C-OS?
> 
> -- Georg

Agreed that there is a "standard" behind XML, but in
the end, after you've specified your document type
definition (or whatever it is now), you still boil
down to conventions and data format.

Beyond that, you are just talking about tools that
work with XML.

I guess what I am saying is that you havn't convinced
me that XML is anything more than a "data format
and standard".
-- 
Warren W. Gay VE3WWG
http://home.cogeco.ca/~ve3wwg

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Georg Bauhaus]

Warren W. Gay VE3WWG <ve3wwg@cogeco.ca> wrote:
: you still boil
: down to conventions and data format.

Sure, and in a sense, an Ada program is just a string
that is in accordance with some grammar, that is
acceepted by some machinery. :-) That is, all this gossip
about the safe programming language Ada boils down
to accepted/not accepted, right?

Besides, is there a convention based data format similar
to XML based formats that provides for builtin referential
integrity (like via IDREF?)? One that has well defined
inclusion and reuse mechanisms? One that allows renaming
of values?

I'm not convinced that calling a BNF like thing (and more)
mere convention is really appropriate.

: Beyond that, you are just talking about tools that
: work with XML.

Well, yes and no (certainly no, actually).  There is a notion
of a conforming XML system, and you can't just specify document
types or even be done with wellformedness and then use, say
pattern based text extraction and say, "we are doing XML based
data processing". At least you'll be missing something then :-)

As a side note, why are these standard lectures called
"algorithms _and_ data structures"? There is even data
driven programming, so a decent data format, and means
of securely specifying it, is crucial, I think.

I'll concede that XML text doesn't look pretty or elegant, but
text with XML markup in it is really useful, for example
when debugging data streams, due to being explicit.
Isn't that something that Ada programmers usually prefer?

-- Georg

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Warren W. Gay VE3WWG]

Georg Bauhaus wrote:
> Warren W. Gay VE3WWG <ve3wwg@cogeco.ca> wrote:
> : you still boil
> : down to conventions and data format.
> 
> Sure, and in a sense, an Ada program is just a string
> that is in accordance with some grammar, that is
> acceepted by some machinery. :-) That is, all this gossip
> about the safe programming language Ada boils down
> to accepted/not accepted, right?

I don't want to get drawn into this, so I'll just focus
on the XML topic at hand.. ;-)

> Besides, is there a convention based data format similar
> to XML based formats that provides for builtin referential
> integrity (like via IDREF?)? One that has well defined
> inclusion and reuse mechanisms? One that allows renaming
> of values?

It doesn't matter how wonderful
the XML format/standard/convention is. To me (at least),
it is _just_ a data format with a standard. :-)

> I'm not convinced that calling a BNF like thing (and more)
> mere convention is really appropriate.

I'm not sure why this is being raised. The original point
was calling an O/S something to do with XML, or to put
"XML" in the name because of its present "hype". I say,
"why do that?" because it is just a data format. Is it
really going to revolutionize the O/S we are talking
about? I guess what I am saying is "not everyone is as
hyped about XML as maybe you are."  Is that a fair
statement to make?

> : Beyond that, you are just talking about tools that
> : work with XML.
> 
> Well, yes and no (certainly no, actually).  There is a notion
> of a conforming XML system, and you can't just specify document
> types or even be done with wellformedness and then use, say
> pattern based text extraction and say, "we are doing XML based
> data processing". At least you'll be missing something then :-)

But _any_ "data format" has a concept of "wellformedness".
This is not unique to XML.

> As a side note, why are these standard lectures called
> "algorithms _and_ data structures"? 

Where does the algorithm come in?  Maybe a poor choice
in title?

> There is even data
> driven programming, so a decent data format, and means
> of securely specifying it, is crucial, I think.

Every program reacts to the input data it is given (ie.
data driven). In fact, in addition to the text and bss
regions, every program has a "data" region. Sounds like
a data driven process to me ;-) . So I fail
to see how XML is "new" in this regard or in this role.
It still just represents plain old data in a new
wrapper.

> I'll concede that XML text doesn't look pretty or elegant, but
> text with XML markup in it is really useful, 

I never said it wasn't useful. ;-)

> for example
> when debugging data streams, due to being explicit.

How do you define "explicit" that is different from
any other data format?  Isn't any defined
format "explicit"?  I don't see XML being special
in this regard.

It is still just a data format ;-)

-- 
Warren W. Gay VE3WWG
http://home.cogeco.ca/~ve3wwg

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Georg Bauhaus]

Warren W. Gay VE3WWG <ve3wwg@cogeco.ca> wrote:
:> I'm not convinced that calling a BNF like thing (and more)
:> mere convention is really appropriate.
: 
: I'm not sure why this is being raised.  The original point
: was calling an O/S something to do with XML, or to put
: "XML" in the name because of its present "hype".

I was reacting to the subject line and other issues that
have been raised, sorry if I'm beeing carried away.
However, BNF is being raised, because one standard isn't
like every other one, they differ in quality and extent;
an "industry standard" might be ``We'll send records,
3 to 5 lines each, where the firsts line is ...''.
Then let's hope that (as is my daily experience) there
is not a line break within one field etc.
According to Goldfarb's law, if there is a bug in some text
processing software, one has to do with line breaks (that is,
naively defined special case data formats :-).

: How do you define "explicit" that is different from
: any other data format?

Here (in this example) the explicitness comes in. Line breaks don't
matter unless you make them correspond to some (empty) element,
or better, forget about line breaks and just mark your fields and
records. In the format described (defined) above you cannot see
the line breaks, in a sense. And if one is missing, or if some
line break sneaks into field content... Likewise, you don't have
to infer where something starts or ends, as you would have to in a
Lisp based format that has recently been suggested. It's familiar:

procedure Rant is
begin
OT: begin
       Don'T("talk about XML in c.l.a");
    end OT;  -- this "named end" can be checked by the compiler!
end Rant;  -- same here

: But _any_ "data format" has a concept of "wellformedness".
: This is not unique to XML.

It isn't, but not any data format has a concept of validity.
And not every format can be checked by a standard toolset,
when there isn't one.

Here is some background, you may conclude why I think that
XML based data streams deserve some hype, no matter what
the marketing press prints. We have to process:
- an ancient video text stream as a news feed
- MS Word xml output (no Word style sheet) with no
  indication of structure but a mix of p-tags and several
  meaningless name spaces in it (job: extract pieces of relevant
  data into some RDB. It's doomed to fail again and again)
- CSV with no line breaks, only group length...

: I guess what I am saying is "not everyone is as
: hyped about XML as maybe you are."  Is that a fair
: statement to make?

I guess so, although I won't call it "hype". XML is roughly
as old as Ada (essentially; it used to be called SGML then).
Data streams are probably at least as old, as are the problems.
XML deserves to be praised, as does Ada when compared to
other options; I've heard that there used to be a lot of
"Ada hype" sometime in the past. The hype doesn't count,
though, does it?

As for operating systems' interfaces, I am rather fond of
Plan 9's, and I think that non-document centric windowsish
window fair grounds are but a big step backwards. Sells well,
though. :-) (Thus, there would be little need to specify the
properties of trappings. Mind you, they have reintroduced
the MDI, and cascading menus! Distructing jugglery, I say.)

:> As a side note, why are these standard lectures called
:> "algorithms _and_ data structures"? 
: 
: Where does the algorithm come in?  Maybe a poor choice
: in title?

Indeed I know of one such lecture that leaves algorithm
out of the title.

: Every program reacts to the input data it is given 
: [...]  So I fail
: to see how XML is "new" in this regard or in this role.
: It still just represents plain old data in a new
: wrapper.

The wrapper isn't new as SGML isn't new. But the wrapper is
the essential thing. Not a T-shirt with holes in it but a
universal jacket helpful in almost all sorts of situations :-)
And the data is not just plain old, it is even plain old text.

: It is still just a data format ;-)

So is BNF? ;-)

-- Georg
---
Microsoft Windows--a fresh perspective on information hiding

RE: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Robert C. Leif]

Yes, however, it is well designed. In order to succeed as a major
language, Ada needs allies.
Bob Leif

-----Original Message-----
From: comp.lang.ada-admin@ada.eu.org
[mailto:comp.lang.ada-admin@ada.eu.org] On Behalf Of Larry Kilgallen
Sent: Monday, September 09, 2002 7:28 AM
To: comp.lang.ada@ada.eu.org
Subject: Re: Advantage of XML based GUI? (was Re: Ada-inspired
OS/Language)

In article <slrnanots5.2b2.randhol+news@kiuk0156.chembio.ntnu.no>,
Preben Randhol <randhol+news@pvv.org> writes:
> On Sun, 8 Sep 2002 15:14:13 -0700, Robert C. Leif wrote:
>> If Nick Roberts et al. wish to generate some real interest and the
>> ability to raise money, call it XMLOS. The GUI would be XML and the
>> insides pure Ada. This is essentially what I suggested in my paper in
>> the last Ada Letters.
> 
> What is the advantage of having the GUI as XML?

I think the point is that XML is trendy.

RE: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Robert C. Leif]

XML forms, XFORMS, can be used to communicate with a programming
language. XFORMS includes data validation, which can include range
checking. The actual XML documents can be Ada protected types, which are
controlled by Ada tasking. The graphical representation can be handled
by either XSL or SVG. What is needed is an Ada implementation of a
permitted subset of XFORMS. This would be sufficient to bootstrap
applications. The trick is to create XML schema with as close to a one
to one correspondence to Ada as possible. The actual implementation of
XFORMS could be based on existing Ada GUI packages.

The commercial virtue of this approach is that one could build a
portable GUI that is based on browsers with the addition of Ada
plug-ins. Basing the GUI on the browser eliminates the problem of
creating CRT and printer device drivers. Thus, one can have Ada software
that will run on Windows and the Mac, as well as run on a native Ada
operating system. 
Although I doubt that it is possible to create an Ada revolution in
software, I believe that it is possible to create an Ada evolution.
Bob Leif

-----Original Message-----
From: comp.lang.ada-admin@ada.eu.org
[mailto:comp.lang.ada-admin@ada.eu.org] On Behalf Of Preben Randhol
Sent: Monday, September 09, 2002 3:27 AM
To: comp.lang.ada@ada.eu.org
Subject: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

On Sun, 8 Sep 2002 15:14:13 -0700, Robert C. Leif wrote:
> If Nick Roberts et al. wish to generate some real interest and the
> ability to raise money, call it XMLOS. The GUI would be XML and the
> insides pure Ada. This is essentially what I suggested in my paper in
> the last Ada Letters.

What is the advantage of having the GUI as XML? I don't get it. Wouldn't
you need a more dynamic language than Ada then?

Preben

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Preben Randhol]

On Mon, 9 Sep 2002 22:21:11 -0700, Robert C. Leif wrote:

[long nice explanation]

so it boils down to two things: GUI portability and XML being in. Costs
are as I see it increase CPU usage and memory usage.

Preben

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Marin David Condic]

But do keep in mind that to a large extent, the price is already being
paid. Any GUI already takes up a bunch of resources just by being a GUI
rather than a command line. When someone builds something like Explorer
or Netscape, they've got to ride on top the OS doing a lot of redundant
work. Doing something where the GUI engine is also basically a browser
means you're consolidating functions and probably the net result is
you're using fewer resources. Isn't Windows already increasingly trying
to use Explorer as the basis for many OS functions? I think the trend is
already started and creating an Ada OS or Ada development/execution
environment that uses XML in some manner as the basis for its GUI is
just a natural extension of what is already happening in the world.

MDC

-- 
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jast.mil/

Send Replies To: m c o n d i c @ a c m . o r g

    "I'd trade it all for just a little more"
        --  Charles Montgomery Burns, [4F10]
======================================================================


Preben Randhol wrote:
> 
> so it boils down to two things: GUI portability and XML being in. Costs
> are as I see it increase CPU usage and memory usage.
>

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Christopher Browne]

After takin a swig o' grog, Marin David Condic <mcondic.auntie.spam@acm.org> belched out...:
> But do keep in mind that to a large extent, the price is already
> being paid. Any GUI already takes up a bunch of resources just by
> being a GUI rather than a command line. When someone builds
> something like Explorer or Netscape, they've got to ride on top the
> OS doing a lot of redundant work. Doing something where the GUI
> engine is also basically a browser means you're consolidating
> functions and probably the net result is you're using fewer
> resources. Isn't Windows already increasingly trying to use Explorer
> as the basis for many OS functions? I think the trend is already
> started and creating an Ada OS or Ada development/execution
> environment that uses XML in some manner as the basis for its GUI is
> just a natural extension of what is already happening in the world.

Let me suggest another possibility, namely that there may be no
significant "price" in this to worry about.

I'll point at GTk's Glade system as an example.

Many (most?) modern applications based on the GNOME application
framework use an XML-based GUI description scheme called Glade.

It can load in, on demand at run time, a fair-sized horde of UI
specifications, including:

 a) A Glade file that specifies the sets of menus that appear at the
    top of the main window;

 b) Many GTK specifications for dialog windows that only get loaded
    occasionally, including:

     advanced-filter.glade col-width.glade exp-smoothing.glade
     mean-tests.glade search-replace.glade analysis-tools.glade
     colrow.glade font-sel.glade moving-averages.glade search.glade
     anova-one.glade consolidate.glade formula-guru.glade names.glade
     sheet-order.glade anova-two.glade correlation.glade
     fourier-analysis.glade plugin-manager.glade sheet-rename.glade
     autocorrect.glade covariance.glade function-select.glade
     print.glade solver.glade autoformat.glade delete-cells.glade
     goalseek.glade random-generation.glade summary.glade
     autosave-prompt.glade descriptive-stats.glade graph-guru.glade
     rank.glade tabulate.glade autosave.glade dialog-stf-export.glade
     hf-config.glade regression.glade variance-tests.glade
     cell-format.glade dialog-stf.glade histogram.glade
     row-height.glade workbook-attr.glade cell-sort.glade
     dialog-zoom.glade insert-cells.glade sampling.glade

With all of these, there is _nothing_ that makes it important that
they be 'compiled' in order to be especially efficient.  They all
represent functions that will get run periodically, and where the user
will _have_ to do some thinking about what to do with the results.  If
it were to take 0.2s to load in the XML file to get the GUI running,
that is unlikely to appreciably injure the "user experience."

Furthermore, for particular UIs that are of particular "performance
importance," nothing prevents you from using some sort of "compiler"
process to transform the XML into native code.

There are compilers for Glade for various languages, including C, C++,
Eiffel, OCaml, Perl, and even (go, figure) Ada95.

You thus generally have a choice:
 a) Load in the XML at runtime, dynamically evaluating and binding it;
 b) Compile it into your code.

Both have advantages and disadvantages:
  -> Compiled code is faster;
  -> Compiled code tends to be bigger (oops);
  -> Compiled code can't be modified at runtime.

The Common Lisp folk have discovered an interesting thing in the realm
of compiling that parallels this; they often like to have both the
ability to compile to bytecode as well as to native machine code.

Native machine code is fastest, and is very valuable for the code that
runs a lot.

On the other hand, bytecode is a LOT more compact.  I found a typical
"factor of five" difference, myself.  Code that doesn't run very often
or in which performance is not critical ought to be bytecode-compiled,
as that provides substantial savings.

There's also another parallel that can be drawn that involves pretty
much the same conclusions:  CORBA.

Most CORBA implementations compile protocol specifications in the IDL
language into native code for the "stubs" and "skeletons" for
marshalling arguments to and from the language's native form into the
form described by the IDL.  In scripting languages, it is not uncommon
for the IDL to dynamically be read at runtime, thus being
'interpreted' in much the same way XML or bytecode would be.

The _exact same results_ fall out.  

If you have a desparate need for marshalling speed, compiled code is
far and away the best way to go.  On the other hand, the compiled
"stubs" and "skeletons" expand into a pretty big blob of compiled
code, so that the speed comes at some cost in memory consumption.
-- 
(reverse (concatenate 'string "moc.enworbbc@" "enworbbc"))
http://cbbrowne.com/info/languages.html
"I'd much rather punch my timecards with a punch awl and blood
spattered fist than use the Solomon IV "Timekeeper" module."
-- Peter da Silva

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[achrist]

See the article on slashdot today pointing to:

http://www.salon.com/tech/feature/2002/09/10/browser_wars/index.html

Which mentions Mozilla as a cross-platform GUI tool that is controlled
by something called XUL.  Maybe an Ada-to-XUL API is what we need.


Al  


Preben Randhol wrote:
> 
> On Mon, 9 Sep 2002 22:21:11 -0700, Robert C. Leif wrote:
> 
> [long nice explanation]
> 
> so it boils down to two things: GUI portability and XML being in. Costs
> are as I see it increase CPU usage and memory usage.
> 
> Preben

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Christopher Browne]

Oops! achrist@easystreet.com was seen spray-painting on a wall:
> See the article on slashdot today pointing to:
>
> http://www.salon.com/tech/feature/2002/09/10/browser_wars/index.html
>
> Which mentions Mozilla as a cross-platform GUI tool that is controlled
> by something called XUL.  Maybe an Ada-to-XUL API is what we need.

Alternatively, <http://libre.act-europe.fr/GtkAda/> is a link to:
_GtkAda: a complete Ada95 graphical toolkit_

UIs are constructed using Glade, which stores UIs in XML form.  They
may be compiled into Ada code using the toolset; there is also support
for loading Glade XML files at runtime.

It is probably a simpler solution to look at something that Already
Exists And Works than to look at something involving pretty gratuitous
amounts of complexity that doesn't work with your favorite tools
yet...
-- 
(reverse (concatenate 'string "moc.enworbbc@" "enworbbc"))
http://cbbrowne.com/info/linux.html
Signs of a Klingon  Programmer #4:  "A  TRUE Klingon Warrior  does not
comment his code!"

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[achrist]

Christopher Browne wrote:
> 
> Oops! achrist@easystreet.com was seen spray-painting on a wall:

What's the excuse for this kind of rudeness?


Al

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Christopher Browne]

Centuries ago, Nostradamus foresaw when achrist@easystreet.com would write:
> Christopher Browne wrote:
>> 
>> Oops! achrist@easystreet.com was seen spray-painting on a wall:
>
> What's the excuse for this kind of rudeness?

(setq *random-headers*
      '(("The world rejoiced as " " wrote")
	("In an attempt to throw the authorities off his trail, " 
	 " transmitted")
	("Oops! "  " was seen spray-painting on a wall")
	("In the last exciting episode, " " wrote:")
	("Quoth " "")
	("A long time ago, in a galaxy far, far away, " " wrote")
	("After takin a swig o' grog, " " belched out...")
	("Centuries ago, Nostradamus foresaw when " " would write")
	("" " wrote")))

(defun insert-random-header ()
  (let* ((r (random (length *random-headers*)))
	 (headers (elt *random-headers* r))
	 (pre (car headers))
	 (post (cadr headers)))
    (insert pre (mail-header-from message-reply-headers) post ":\n")))

At least you didn't get the "swig o' grog" one...
-- 
(concatenate 'string "aa454" "@freenet.carleton.ca")
http://cbbrowne.com/info/finances.html
Pagers are cases for holding dead batteries. -Richard Wolff 

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Ted Dennison]

Preben Randhol <randhol+news@pvv.org> wrote in message news:<slrnanots5.2b2.randhol+news@kiuk0156.chembio.ntnu.no>...
> On Sun, 8 Sep 2002 15:14:13 -0700, Robert C. Leif wrote:
> > If Nick Roberts et al. wish to generate some real interest and the
> > ability to raise money, call it XMLOS. The GUI would be XML and the
> > insides pure Ada. This is essentially what I suggested in my paper in
> > the last Ada Letters.
> 
> What is the advantage of having the GUI as XML? I don't get it. Wouldn't
> you need a more dynamic language than Ada then?


I'd suggest anyone interested in this issue head over to
http://www.xulplanet.com/tutorials/xultu/ , and take a look at exactly
how this works and what can be done with it.

I haven't done so myself yet, so I'll resist the urge to post an
opinion of my own here.

RE: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Robert C. Leif]

XUL has some good ideas; however, it is a bit of a hack. It uses RDF,
which is a "concise" replacement for schemas. The XUL model could be
followed with a limited implementation of XFORMS. The programming
language should be Ada. 
I believe that cascading style sheets are obsolete. A simple version of
XSL, SVG or even the use of an Ada application, such as CLAW would be
sufficient to get the system started. 
I would like to suggest one design rule, Less is better! The average
customer would have an increased feeling of security with a very simple
system. The great advantage of DOS was that it was NOT an operating
system. The PC is more of an appliance than a computer. I suspect that
Ada technology may be able to provide a system where one can pull the
plug and then virtually instantly reboot.
Bob Leif

-----Original Message-----
From: comp.lang.ada-admin@ada.eu.org
[mailto:comp.lang.ada-admin@ada.eu.org] On Behalf Of Ted Dennison
Sent: Tuesday, September 10, 2002 7:29 PM
To: comp.lang.ada@ada.eu.org
Subject: Re: Advantage of XML based GUI? (was Re: Ada-inspired
OS/Language)

Preben Randhol <randhol+news@pvv.org> wrote in message
news:<slrnanots5.2b2.randhol+news@kiuk0156.chembio.ntnu.no>...
> On Sun, 8 Sep 2002 15:14:13 -0700, Robert C. Leif wrote:
> > If Nick Roberts et al. wish to generate some real interest and the
> > ability to raise money, call it XMLOS. The GUI would be XML and the
> > insides pure Ada. This is essentially what I suggested in my paper
in
> > the last Ada Letters.
> 
> What is the advantage of having the GUI as XML? I don't get it.
Wouldn't
> you need a more dynamic language than Ada then?


I'd suggest anyone interested in this issue head over to
http://www.xulplanet.com/tutorials/xultu/ , and take a look at exactly
how this works and what can be done with it.

I haven't done so myself yet, so I'll resist the urge to post an
opinion of my own here.

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Programmer Dude]

"Robert C. Leif" wrote:

> I suspect that Ada technology may be able to provide a system where
> one can pull the plug and then virtually instantly reboot.

That would be delightful!  It's always seemed ironic to me that the
faster PCs get, the long they take to boot.

BTW, Hello!  I'm a long time programmer (25 years behind the keyboard,
boy and man) who's been recently becoming increasingly interested in
Ada.  I've been lurking here, reading Ada Distilled and have downloaded
GNAT and some other goodies.

I'll go back to lurking for now, but I just wanted to say, "Hi!"

-- 
|_ CJSonnack <Chris@Sonnack.com> _____________| How's my programming? |
|_ http://www.Sonnack.com/ ___________________| Call: 1-800-DEV-NULL  |
|_____________________________________________|_______________________|



Opinions expressed herein are my own and may not represent those of my employer.

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Nick Roberts]

On 10 Sep 2002 19:28:43 -0700, dennison@telepath.com (Ted Dennison)
strongly typed:

>I'd suggest anyone interested in this issue head over to
>http://www.xulplanet.com/tutorials/xultu/ , and take a look at exactly
>how this works and what can be done with it.

Very like as I envisaged (an XML based RAD language). I would like to get
something like this working on AdaOS as early as would be feasible; I do
think it would be a great way of getting very many useful, practical
applications up and working very rapidly.

-- 
Nick Roberts

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Preben Randhol]

On Wed, 11 Sep 2002 19:48:28 GMT, Nick Roberts wrote:
> Very like as I envisaged (an XML based RAD language). I would like to get
> something like this working on AdaOS as early as would be feasible; I do
> think it would be a great way of getting very many useful, practical
> applications up and working very rapidly.

I have started to read a bit on AdaOS and I have two questions. When it
comes to device drivers how are they going to be implemented. I mean one
of the problems Linux has is that one have to make all the device
drivers onself and somethimes this means reverse engineering. Would one
need to do the same for AdaOS or could on easily port the linux
drivers?

Second:

   Printout Renderer

   This program accepts a data stream in the form of binary 'formatting
   objects', and transforms them into the data to be sent to the printer
   (e.g. PostScript, PCL, or pure TIFF).

   One of the special extra functions of this program is to separately
   accept text for a page header and page footer, which it prints on
   every page. This function is used to ensure that security labels are
   printed.

I don't understand the latter paragraph. Say I want to print a book or
something else from my AdaOS, I wouldn't want extra headers and
footnotes to clutter the page so I guess one can turn this off? But if
on what does it do to security?

Preben Randhol

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Chad R. Meiners]

"Preben Randhol" <randhol+news@pvv.org> wrote in message
news:slrnao0ihd.14r.randhol+news@kiuk0156.chembio.ntnu.no...
> I don't understand the latter paragraph. Say I want to print a book or
> something else from my AdaOS, I wouldn't want extra headers and
> footnotes to clutter the page so I guess one can turn this off? But if
> on what does it do to security?

For ensuring that "TOP SECRET" is printed on every page of the top secret
document you are printing.

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Andrew Berg]

"Chad R. Meiners" <crmeiners@hotmail.com> wrote in message news:<alq8bs$9ck$1@msunews.cl.msu.edu>...
> "Preben Randhol" <randhol+news@pvv.org> wrote in message
> news:slrnao0ihd.14r.randhol+news@kiuk0156.chembio.ntnu.no...
> > I don't understand the latter paragraph. Say I want to print a book or
> > something else from my AdaOS, I wouldn't want extra headers and
> > footnotes to clutter the page so I guess one can turn this off? But if
> > on what does it do to security?
> 
> For ensuring that "TOP SECRET" is printed on every page of the top secret
> document you are printing.

Haven't your read about M$'s Palladium?  We'll ensure privacy by not
allowing you to print it in the first place!  Won't let you copy, view
or edit files either.

-andrew

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Chad R. Meiners]

So what does your statement have to do with my comment?

"Andrew Berg" <andrew_c_berg@yahoo.com> wrote in message
news:acdd9439.0209180949.4134176f@posting.google.com...
> Haven't your read about M$'s Palladium?  We'll ensure privacy by not
> allowing you to print it in the first place!  Won't let you copy, view
> or edit files either.
>
> -andrew

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Nick Roberts]

On Thu, 12 Sep 2002 08:02:53 +0000 (UTC), Preben Randhol
<randhol+news@pvv.org> strongly typed:

>On Wed, 11 Sep 2002 19:48:28 GMT, Nick Roberts wrote:
>> Very like as I envisaged (an XML based RAD language). I would like to get
>> something like this working on AdaOS as early as would be feasible; I do
>> think it would be a great way of getting very many useful, practical
>> applications up and working very rapidly.
>
>I have started to read a bit on AdaOS and I have two questions. When it
>comes to device drivers how are they going to be implemented. I mean one
>of the problems Linux has is that one have to make all the device
>drivers onself and somethimes this means reverse engineering. Would one
>need to do the same for AdaOS or could on easily port the linux
>drivers?

Currently, I plan to write 'native' device drivers (in Ada) for AdaOS. I
intend to go to the original hardware documentation wherever posisble, and
write the drivers properly. I don't think it's unfair to suggest that most
device drivers (even commercial ones) tend be of remarkably poor quality.

Nevertheless, it would be really handy to provide compatibility layers that
would allow existing device drivers to be used. At some stage, maybe we
will have a go at this.

>Second:
>
>   Printout Renderer
>
>   This program accepts a data stream in the form of binary 'formatting
>   objects', and transforms them into the data to be sent to the printer
>   (e.g. PostScript, PCL, or pure TIFF).
>
>   One of the special extra functions of this program is to separately
>   accept text for a page header and page footer, which it prints on
>   every page. This function is used to ensure that security labels are
>   printed.
>
>I don't understand the latter paragraph. Say I want to print a book or
>something else from my AdaOS, I wouldn't want extra headers and
>footnotes to clutter the page so I guess one can turn this off? But if
>on what does it do to security?

Chad Meiners' answer:

| For ensuring that "TOP SECRET" is printed on every page of the top secret
| document you are printing.

is absolutely correct (in essence).

If you don't want labels to be printed, and you have the necessary security
clearance, you simply tell Printout Manager not to print them (it will be
an option). The point is that this is a privileged option; the system
administrator(s) can stop ordinary users using this option, if they want
to.

All security features are optional. It's just that they are optional for
the privileged user, not for the ordinary user!

If you were running AdaOS on your own personal computer, you would be the
system administrator, and so you would have the highest privileges (you are
allowed to do anything ;-). Probably you would have security label printing
permanently turned off. In fact, probably you would have no mandatory
security features operative at all.

If, on the other hand, you were using a workstation which is a part of your
company's AdaOS network, you may be permitted to print things, but not to
turn off the security labels. If this was going to be a problem, you would
have to ask a privileged user to print it for you (with the labels off);
that user would have to take responsibility for the possible consequences.

-- 
Nick Roberts

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Preben Randhol]

On Thu, 12 Sep 2002 16:12:12 GMT, Nick Roberts wrote:
> All security features are optional. It's just that they are optional for
> the privileged user, not for the ordinary user!

I don't think I understand what is inherently more secure that it says
Top Secret on the top of some piece of paper. To me it sounds like
something only usable in the military?

> If you were running AdaOS on your own personal computer, you would be the
> system administrator, and so you would have the highest privileges (you are
> allowed to do anything ;-). Probably you would have security label printing
> permanently turned off. In fact, probably you would have no mandatory
> security features operative at all.

I hope you are thinking of doing it the UNIX way. I mean that the user
does not have root/administrator privileges and that there is only one
root superuser which is not used unless when one needs to install
software.

This document should be of interest: http://www.nsa.gov/selinux/

Preben Randhol

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[David C. Hoos]

----- Original Message -----
From: "Preben Randhol" <randhol+news@pvv.org>
Newsgroups: comp.lang.ada
To: <comp.lang.ada@ada.eu.org>
Sent: Thursday, September 12, 2002 11:45 AM
Subject: Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)


> On Thu, 12 Sep 2002 16:12:12 GMT, Nick Roberts wrote:
> > All security features are optional. It's just that they are optional for
> > the privileged user, not for the ordinary user!
>
> I don't think I understand what is inherently more secure that it says
> Top Secret on the top of some piece of paper. To me it sounds like
> something only usable in the military?
>
It's just to enforce that a secret document cannot be printed without each
of
its pages being labelled as "secret," etc. I.e., _every_ document printed on
a printer accessible from computers with classified material _must_ print
a security label on each page -- even if it's only to say "UNCLASSIFIED"

<snip>

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Preben Randhol]

On Thu, 12 Sep 2002 12:03:55 -0500, David C. Hoos wrote:
> It's just to enforce that a secret document cannot be printed without each
> of
> its pages being labelled as "secret," etc. I.e., _every_ document printed on
> a printer accessible from computers with classified material _must_ print
> a security label on each page -- even if it's only to say "UNCLASSIFIED"

If you are in the military I guess, but not for non military use.

But if you then have two PS files. One that is Top Secret and one that
is Unrestricted (or what it is called), how will the system know to put
the right labels on them? Will the files have a security level attribute?

Preben

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Nick Roberts]

On Fri, 13 Sep 2002 07:57:02 +0000 (UTC), Preben Randhol
<randhol+news@pvv.org> strongly typed:

>On Thu, 12 Sep 2002 12:03:55 -0500, David C. Hoos wrote:
>> It's just to enforce that a secret document cannot be printed without each
>> of
>> its pages being labelled as "secret," etc. I.e., _every_ document printed on
>> a printer accessible from computers with classified material _must_ print
>> a security label on each page -- even if it's only to say "UNCLASSIFIED"
>
>If you are in the military I guess, but not for non military use.
>
>But if you then have two PS files. One that is Top Secret and one that
>is Unrestricted (or what it is called), how will the system know to put
>the right labels on them? Will the files have a security level attribute?

Are you asking about AdaOS, Preben?

If so, the answer is 'yes'. Every single data object in the system will
have two 'mandatory' security labels: a 'confidentiality label' and an
'integrity label'. The system will enforce the following basic semantics:
no information is allowed to flow from a data object to another data object
with a lower confidentiality label; no information is allowed to flow from
a data object to another data object with a higher integrity label.

In AdaOS, everything will be an object. A file will be an object.
Furthermore, it'll be a data object, so it will have mandatory labels.

So, if a file "Next Generation Stealth Technology 2002.ps" has
confidentiality label "Top Secret", and I am (represented by) an object
with the lower confidentiality label "Mere Pion", I will be prevented from
accessing the data in the report. Nor will it be permitted to print the
file on the printer in the office lobby (with confidentiality label "Anyone
Can See This"); it must be printed on a printer with a confidentiality
label equal to or higher than "Top Secret", and even then it will have "TOP
SECRET" printed on the top and bottom of every page. This all assumes an
organisation that has set up its computer security properly.

And I say 'organisation', because I could be talking about a DoD
department, or I could just as easily be talking about Lockheed Martin
(Skunkworks) or some other corporation. Consider the banks and other
financial institutions; isn't it vital that their data is protected? Then
consider medical records. Passport, national insurance, and immigration
records. Police and court records. The list goes on and on. Their data
security is vital.

-- 
Nick Roberts

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Marin David Condic]

Proividing a superior level of security management is another area that
could greatly differentiate AdaOS from other systems. This would help create
a niche in which it could gain acceptance and gradually find applications
getting ported to it. That overcomes a lot because its obvious that any new
OS has to play "Catch Up" with the others in terms of getting apps ported to
it. Here, you'd be able to say "Well, maybe we don't have the latest version
of Space Invaders or Auto-CAD running on AdaOS, but we've got at least the
common Office tools and if you need security around your documents, this is
the way to go..."

MDC
--
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jast.mil/

Send Replies To: m c o n d i c @ a c m . o r g

    "I'd trade it all for just a little more"
        --  Charles Montgomery Burns, [4F10]
======================================================================

Nick Roberts <nickroberts@blueyonder.co.uk> wrote in message
news:3d822b59.1088140942@news.cis.dfn.de...
>
> And I say 'organisation', because I could be talking about a DoD
> department, or I could just as easily be talking about Lockheed Martin
> (Skunkworks) or some other corporation. Consider the banks and other
> financial institutions; isn't it vital that their data is protected? Then
> consider medical records. Passport, national insurance, and immigration
> records. Police and court records. The list goes on and on. Their data
> security is vital.
>

RE: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Robert C. Leif]

The common Office Tools could be done as XML applications based on
schemas and XSL. I presently use Adobe FrameMaker for all of my serious
writing and our office still uses Lotus Improv. Improv, which started
with NextStep and migrated to Windows was killed by Lotus. It has the
great advantage of being able to rename the rows and columns with unique
names. Therefore, the formulas are human readable. FrameMaker has the
great advantage that it lacks Word's artificial stupidity and thus,
maintains its paragraph definitions. Parenthetically, the spreadsheet
should serve as the tables in the wordprocessor. 
Bob Leif

-----Original Message-----
From: comp.lang.ada-admin@ada.eu.org
[mailto:comp.lang.ada-admin@ada.eu.org] On Behalf Of Marin David Condic
Sent: Saturday, September 14, 2002 6:00 AM
To: comp.lang.ada@ada.eu.org
Subject: Re: Advantage of XML based GUI? (was Re: Ada-inspired
OS/Language)

Proividing a superior level of security management is another area that
could greatly differentiate AdaOS from other systems. This would help
create
a niche in which it could gain acceptance and gradually find
applications
getting ported to it. That overcomes a lot because its obvious that any
new
OS has to play "Catch Up" with the others in terms of getting apps
ported to
it. Here, you'd be able to say "Well, maybe we don't have the latest
version
of Space Invaders or Auto-CAD running on AdaOS, but we've got at least
the
common Office tools and if you need security around your documents, this
is
the way to go..."

MDC
--
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jast.mil/

Send Replies To: m c o n d i c @ a c m . o r g

    "I'd trade it all for just a little more"
        --  Charles Montgomery Burns, [4F10]
======================================================================

Nick Roberts <nickroberts@blueyonder.co.uk> wrote in message
news:3d822b59.1088140942@news.cis.dfn.de...
>
> And I say 'organisation', because I could be talking about a DoD
> department, or I could just as easily be talking about Lockheed Martin
> (Skunkworks) or some other corporation. Consider the banks and other
> financial institutions; isn't it vital that their data is protected?
Then
> consider medical records. Passport, national insurance, and
immigration
> records. Police and court records. The list goes on and on. Their data
> security is vital.
>

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Preben Randhol]

On Sat, 14 Sep 2002 08:59:52 -0400, Marin David Condic wrote:
> Proividing a superior level of security management is another area that
> could greatly differentiate AdaOS from other systems. This would help create
> a niche in which it could gain acceptance and gradually find applications
> getting ported to it. That overcomes a lot because its obvious that any new
> OS has to play "Catch Up" with the others in terms of getting apps ported to
> it. Here, you'd be able to say "Well, maybe we don't have the latest version
> of Space Invaders or Auto-CAD running on AdaOS, but we've got at least the
> common Office tools and if you need security around your documents, this is
> the way to go..."

I guess the server-market is easier to target first. Say a secure AdaOS
runnig a f.ex AWS server. That would have been fun.

Preben

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Marin David Condic]

Thats how Linux got some traction. A server doesn't necessarily have to have
thousands of diverse apps available to be usable. Nick's notion of making
the OS highly distributed muight lend itself to this - a scalable server
with high security.

MDC
--
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jast.mil/

Send Replies To: m c o n d i c @ a c m . o r g

    "I'd trade it all for just a little more"
        --  Charles Montgomery Burns, [4F10]
======================================================================

Preben Randhol <randhol+abuse@pvv.org> wrote in message
news:slrnaob4nb.o7.randhol+abuse@kiuk0152.chembio.ntnu.no...
>
> I guess the server-market is easier to target first. Say a secure AdaOS
> runnig a f.ex AWS server. That would have been fun.
>

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Preben Randhol]

On Mon, 16 Sep 2002 07:47:33 -0400, Marin David Condic wrote:
> Thats how Linux got some traction. A server doesn't necessarily have to have
> thousands of diverse apps available to be usable. Nick's notion of making
> the OS highly distributed muight lend itself to this - a scalable server
> with high security.

Yes. This should be great for clustering shouldn't it? I don't know so
much about it, but can't you use the distributedness to easily do
calculations on several machines at once? So one could reboot into AdaOS
before one goes home at night and leave the computer to do calculations
all night while you sleep without worrying too much that somebody is
breaking into your system.

Preben who always turns off his office machine, firewalls or not.

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Marin David Condic]

Well, yes, but I don't think theres any way of getting it free of charge. If
the program you have is not designed to be distributed, it would be really
tough for the OS to figure out what parts of it to distribute. If you broke
it up into tasks, then obviously, the OS stands a reasonable chance of
distributing the tasks. If you've gone that far in designing your
application, you can clearly find a way to make it distributed through your
own manual methods - but it would be nice for the OS to just do it for you.

I recall that at one point in time, Pratt & Whitney designed a distributed
application for doing computational fluid dynamics. At night, all the Sun
workstations they had could be put to work solving problems that used to
require quite a bit of time on a Cray supercomputer. When they got it
working, they fired the Cray and just used cheap Sun labor. It worked pretty
well.

MDC
--
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jast.mil/

Send Replies To: m c o n d i c @ a c m . o r g

    "I'd trade it all for just a little more"
        --  Charles Montgomery Burns, [4F10]
======================================================================

Preben Randhol <randhol+news@pvv.org> wrote in message
news:slrnaobid8.5e7.randhol+news@kiuk0156.chembio.ntnu.no...
>
> Yes. This should be great for clustering shouldn't it? I don't know so
> much about it, but can't you use the distributedness to easily do
> calculations on several machines at once? So one could reboot into AdaOS
> before one goes home at night and leave the computer to do calculations
> all night while you sleep without worrying too much that somebody is
> breaking into your system.
>
> Preben who always turns off his office machine, firewalls or not.

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Georg Bauhaus]

Preben Randhol <randhol+news@pvv.org> wrote:
: I hope you are thinking of doing it the UNIX way.

Hm. Even the Unix guys have made something else:
http://plan9.bell-labs.com/plan9dist/index.html


-- Georg

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Preben Randhol]

On Thu, 12 Sep 2002 18:57:42 +0000 (UTC), Georg Bauhaus wrote:
> Preben Randhol <randhol+news@pvv.org> wrote:
>: I hope you are thinking of doing it the UNIX way.
> 
> Hm. Even the Unix guys have made something else:
> http://plan9.bell-labs.com/plan9dist/index.html

I don't know Plan 9 although the screenshot looks quite like UNIX. But
this is beside my point. I was only refering to how the administrator is
going to be delt with. Perhaps plan 9 has a nice system for this, but
could you explain how they do it?

Preben

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Georg Bauhaus]

Preben Randhol <randhol+news@pvv.org> wrote:
: going to be delt with. Perhaps plan 9 has a nice system for this, but
: could you explain how they do it?

There is not really a Unix root user, though there are privileges.
(There are file servers, compute servers, and workstations,
not necessarily tied to machines.)
User rights are elaborate, distributed, sometimes surprinsingly absent,
if you want this, the file system (which is central to everything)
has namespaces such that you can import "parts" of some "file area"
into some other "file area"...

Please don't rely on this description, I wish I had more time
to learn more about this system. It feels really different.
Productive. Simple. Good.

--  Georg
---
Microsoft Windows--a fresh perspective on information hiding

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Preben Randhol]

On Fri, 13 Sep 2002 12:23:59 +0000 (UTC), Georg Bauhaus wrote:
> There is not really a Unix root user, though there are privileges.
> (There are file servers, compute servers, and workstations,
> not necessarily tied to machines.)
> User rights are elaborate, distributed, sometimes surprinsingly absent,
> if you want this, the file system (which is central to everything)
> has namespaces such that you can import "parts" of some "file area"
> into some other "file area"...
> 
> Please don't rely on this description, I wish I had more time
> to learn more about this system. It feels really different.
> Productive. Simple. Good.

Hmm sounds interesting.

> Microsoft Windows--a fresh perspective on information hiding

Unless you get a virus that is only so happy to share your information
with the rest of the net.

Preben

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Nick Roberts]

On Thu, 12 Sep 2002 16:45:14 +0000 (UTC), Preben Randhol
<randhol+news@pvv.org> strongly typed:

>On Thu, 12 Sep 2002 16:12:12 GMT, Nick Roberts wrote:
>> All security features are optional. It's just that they are optional for
>> the privileged user, not for the ordinary user!
>
>I don't think I understand what is inherently more secure that it says
>Top Secret on the top of some piece of paper. To me it sounds like
>something only usable in the military?

It is a requirement specified in the TCSEC; that is my primary concern.

I presume the rationale is that it may make it more difficult for someone
to pass off a printed document as having a status it does not. Often, in
the real world, security is about making things difficult (rather than
completely impossible).

This kind of security 'labelling' is a typical military requirement, but
(the TCSEC makes it clear) it would be likely to also be a requirement of a
great many organisations, were they to undertake a true security
requirements assessment. A great many organisations need to keep certain
information confidential, and ensure certain information is correct, as a
legal requirement; most corporations have industrial secrets to protect and
all need to prevent fraud. Of course, the unfortunate fact is that most
organisations today are hopelessly deficient in their overall security
provisions.

>> If you were running AdaOS on your own personal computer, you would be the
>> system administrator, and so you would have the highest privileges (you are
>> allowed to do anything ;-). Probably you would have security label printing
>> permanently turned off. In fact, probably you would have no mandatory
>> security features operative at all.
>
>I hope you are thinking of doing it the UNIX way. I mean that the user
>does not have root/administrator privileges and that there is only one
>root superuser which is not used unless when one needs to install
>software.

AdaOS will be a fully object oriented OS, and its design is not closely
related to any of the Unixen. I am considering the various security schemes
offered by CORBA at the moment. I may well choose to enhance these with
many further facilities. 

>This document should be of interest: http://www.nsa.gov/selinux/

Interesting, but there's nothing new to me here. AdaOS will be designed to
conform to the requirements of the TCSEC level B2 or B1. This exceeds what
is offered by SE Linux (or any of the BSDs).

The security mechanisms of AdaOS will be built into each object (class),
and will be tailored to the specific needs of that object (class); this is
a considerably more sophisticated approach than any isomorphic OS can
offer.

-- 
Nick Roberts

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Preben Randhol]

On Fri, 13 Sep 2002 01:53:42 GMT, Nick Roberts wrote:
>>I hope you are thinking of doing it the UNIX way. I mean that the user
>>does not have root/administrator privileges and that there is only one
>>root superuser which is not used unless when one needs to install
>>software.
> 
> AdaOS will be a fully object oriented OS, and its design is not closely
> related to any of the Unixen. I am considering the various security schemes
> offered by CORBA at the moment. I may well choose to enhance these with
> many further facilities. 

I was only thinking in the respect of how you define users.

> The security mechanisms of AdaOS will be built into each object (class),
> and will be tailored to the specific needs of that object (class); this is
> a considerably more sophisticated approach than any isomorphic OS can
> offer.

So if somebody runs a program that contains a virus of some kind he will
not deleting the hard disc for example? The reason for my question was
that one can in Linux choose to login as root all the time when one has
ones own machine, but this is utterly stupid. The reason is that if you
run a rouge program it will have access to all the disc and can do
anything as it would be run under superuser priveliges. Therefore one
makes a normal user and uses this all the time and only use the
superuser when one need to install software. In Windows they put the
superuser priveliges on one of the users... That's why I'm asking.

Preben

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Nick Roberts]

On Fri, 13 Sep 2002 08:24:44 +0000 (UTC), Preben Randhol
<randhol+news@pvv.org> strongly typed:

>On Fri, 13 Sep 2002 01:53:42 GMT, Nick Roberts wrote:
>>>I hope you are thinking of doing it the UNIX way. I mean that the user
>>>does not have root/administrator privileges and that there is only one
>>>root superuser which is not used unless when one needs to install
>>>software.
>> 
>> AdaOS will be a fully object oriented OS, and its design is not closely
>> related to any of the Unixen. I am considering the various security schemes
>> offered by CORBA at the moment. I may well choose to enhance these with
>> many further facilities. 
>
>I was only thinking in the respect of how you define users.
>
>> The security mechanisms of AdaOS will be built into each object (class),
>> and will be tailored to the specific needs of that object (class); this is
>> a considerably more sophisticated approach than any isomorphic OS can
>> offer.
>
>So if somebody runs a program that contains a virus of some kind he will
>not deleting the hard disc for example? The reason for my question was
>that one can in Linux choose to login as root all the time when one has
>ones own machine, but this is utterly stupid. The reason is that if you
>run a rouge program it will have access to all the disc and can do
>anything as it would be run under superuser priveliges. Therefore one
>makes a normal user and uses this all the time and only use the
>superuser when one need to install software. In Windows they put the
>superuser priveliges on one of the users... That's why I'm asking.

In simple terms, we will take steps to ensure that this sort of this cannot
happen in standard AdaOS installations.

For example, when a program is downloaded from the Internet and executed,
it will be executed with an 'authority' that is very limited (certainly
disallowing reformatting the hard disk ;-)  If that program tries to access
an object it (it's authority) does not have permission to access, the user
will be asked whether to grant that access; the user can say "yes" (in
which case the permission is effectively added to the authority) or "no".
Objects will have security 'groups', to make this process a little less
laborious.

In AdaOS, when a user logs on, the authentication program (called "Quis")
grants him or her a set of authorities, by the expedient of running a
representative (avatar) program, such as "Jumbo" (GUI Session Manager),
which has been granted those authorities. It is these authorities which
carry security rights and information. Whenever the program accesses
another object, it cites an authority; it is this authority that is used by
the object to check the validity of the access (and in any auditing log it
may make).

Each machine will have one special user, the 'workstation administrator',
who is granted an authority that permits full access to every object in or
created by the workstation. This will include, for example, the hard disk
(direct access to it).

However, every user will also be able (and encouraged) to define multiple
'roles' for himself or herself. Each role will define which authorities to
use in conjunction with which objects and which types of access, and a set
of authorities to grant. When the user runs a program, they specify a role,
and this controls the authorities granted to the program, and the default
ways in which it uses them to access objects.

Thus, a user could have a role "Workstation Administrator", as well as
other roles such as "Idle Web Wanderer". "Workstation Administrator" may
grant the authority that permits reformatting of the hard disk. Probably
"Idle Web Wanderer" will not. So, by citing the role "Workstation
Administrator" only when necessary, the user can help protect himself from
doing something disastrous.

-- 
Nick Roberts

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Warren W. Gay VE3WWG]

Nick Roberts wrote:
> On Fri, 13 Sep 2002 08:24:44 +0000 (UTC), Preben Randhol
> <randhol+news@pvv.org> strongly typed:
>>On Fri, 13 Sep 2002 01:53:42 GMT, Nick Roberts wrote:
...
>>So if somebody runs a program that contains a virus of some kind he will
>>not deleting the hard disc for example? The reason for my question was
>>that one can in Linux choose to login as root all the time when one has
>>ones own machine, but this is utterly stupid. The reason is that if you
>>run a rouge program it will have access to all the disc and can do
>>anything as it would be run under superuser priveliges. Therefore one
>>makes a normal user and uses this all the time and only use the
>>superuser when one need to install software. In Windows they put the
>>superuser priveliges on one of the users... That's why I'm asking.
> 
> In simple terms, we will take steps to ensure that this sort of this cannot
> happen in standard AdaOS installations.
> 
> For example, when a program is downloaded from the Internet and executed,
> it will be executed with an 'authority' that is very limited (certainly
> disallowing reformatting the hard disk ;-)  If that program tries to access
> an object it (it's authority) does not have permission to access, the user
> will be asked whether to grant that access; the user can say "yes" (in
> which case the permission is effectively added to the authority) or "no".
> Objects will have security 'groups', to make this process a little less
> laborious.

This doesn't always work out well with the "user". This is the same problem
with the McAfee Firewall software. They start with the premise of locking
everything down (which is good) and then asking questions of the user
as they progress through their normal tasks.

The problem with this model is that the end user
does'nt understand the questions! If the user is presented with a prompt
to allow/disallow TCP port 23 incoming access, he is stuck. The user becomes
impatient and will then:

   1) Allow the access to see if it will accomplish his task (but here he
      has no clue as to how to undo this later)
   2) Disallow any access he doesn't understand because he is paranoid, and
      as a result may not accomplish what he wants to do.
   3) Or answer a combination of 1 & 2 depending upon whether it "sounds
      dangerous" or not.
   4) Do nothing without help, because he is paralyzed with fear.

For McAffee Firewall, it would be better for a knowledgeable friend to set
it up ahead of time so that the end user does not have to decide (at least
it should have that option).

In an operating system situation, I'm not sure if asking the user is the
right thing to do. A knowledgable person should establish the run-time
criteria ahead of time, IMHO.  You don't want end users having to decide
"programming" decisions. The end users don't want that responsibility either.

-- 
Warren W. Gay VE3WWG
http://home.cogeco.ca/~ve3wwg

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Chad R. Meiners]

"Warren W. Gay VE3WWG" <ve3wwg@cogeco.ca> wrote in message
news:3D8358E1.9080703@cogeco.ca...
> This doesn't always work out well with the "user". This is the same
problem
> with the McAfee Firewall software. They start with the premise of locking
> everything down (which is good) and then asking questions of the user
> as they progress through their normal tasks.
>
> The problem with this model is that the end user
> does'nt understand the questions! If the user is presented with a prompt
> to allow/disallow TCP port 23 incoming access, he is stuck. The user
becomes
> impatient and will then:

Agreed, I have given some thought about this.  In my view all programs
should come will installation information.  This information must be
submitted to the OS at install time before it can be considered executable
by anyone.  Each installation lists what capabilities the program needs and
should be accompained with a plain text description of why the program needs
that capablility.  The operating system can also maintain a list of
capability descriptions so the user can compare the needs with the expected
usage.  Based on this information the user can grant the program
all/some/none of the requested capabilities.  If the user wants to install a
program that does not have the accompanied installation information, they
could run it in the interactive mode Nick described.

-CRM

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Nick Roberts]

On Sat, 14 Sep 2002 12:19:57 -0400, "Chad R. Meiners"
<crmeiners@hotmail.com> strongly typed:

>
>"Warren W. Gay VE3WWG" <ve3wwg@cogeco.ca> wrote in message
>news:3D8358E1.9080703@cogeco.ca...
>> This doesn't always work out well with the "user". This is the same
>> problem with the McAfee Firewall software. They start with the
>> premise of locking everything down (which is good) and then asking 
>> questions of the user as they progress through their normal tasks.
>>
>> The problem with this model is that the end user
>> does'nt understand the questions! If the user is presented with a prompt
>> to allow/disallow TCP port 23 incoming access, he is stuck. The user
>> becomes impatient and will then:
>> ...

>Agreed, 

Also agreed.

>I have given some thought about this.  In my view all programs
>should come will installation information.  This information must be
>submitted to the OS at install time before it can be considered executable
>by anyone.  Each installation lists what capabilities the program needs and
>should be accompained with a plain text description of why the program needs
>that capablility.  The operating system can also maintain a list of
>capability descriptions so the user can compare the needs with the expected
>usage.  Based on this information the user can grant the program
>all/some/none of the requested capabilities.  If the user wants to install a
>program that does not have the accompanied installation information, they
>could run it in the interactive mode Nick described.

Perfect. This is indeed (honest!) just how I intend to do it (I thought
about installation matters a lot earlier this year).

In addition, the system will keep track of what was installed from where,
permitting intelligent backup, automatic piecemeal installation and update
from the Internet, and all manner of nice things. All done with careful
consideration of security (including viruses).

-- 
Nick Roberts
Per Ardua ad Disastra

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Mark Biggar]

Nick Roberts wrote:
> On Thu, 12 Sep 2002 16:45:14 +0000 (UTC), Preben Randhol
> <randhol+news@pvv.org> strongly typed:
> 
> 
>>On Thu, 12 Sep 2002 16:12:12 GMT, Nick Roberts wrote:
>>
>>>All security features are optional. It's just that they are optional for
>>>the privileged user, not for the ordinary user!
>>
>>I don't think I understand what is inherently more secure that it says
>>Top Secret on the top of some piece of paper. To me it sounds like
>>something only usable in the military?
> 
> 
> It is a requirement specified in the TCSEC; that is my primary concern.
> 
> I presume the rationale is that it may make it more difficult for someone
> to pass off a printed document as having a status it does not. Often, in
> the real world, security is about making things difficult (rather than
> completely impossible).
> 
> This kind of security 'labelling' is a typical military requirement, but
> (the TCSEC makes it clear) it would be likely to also be a requirement of a
> great many organisations, were they to undertake a true security
> requirements assessment. A great many organisations need to keep certain
> information confidential, and ensure certain information is correct, as a
> legal requirement; most corporations have industrial secrets to protect and
> all need to prevent fraud. Of course, the unfortunate fact is that most
> organisations today are hopelessly deficient in their overall security
> provisions.
> 
> 
>>>If you were running AdaOS on your own personal computer, you would be the
>>>system administrator, and so you would have the highest privileges (you are
>>>allowed to do anything ;-). Probably you would have security label printing
>>>permanently turned off. In fact, probably you would have no mandatory
>>>security features operative at all.
>>
>>I hope you are thinking of doing it the UNIX way. I mean that the user
>>does not have root/administrator privileges and that there is only one
>>root superuser which is not used unless when one needs to install
>>software.
> 
> 
> AdaOS will be a fully object oriented OS, and its design is not closely
> related to any of the Unixen. I am considering the various security schemes
> offered by CORBA at the moment. I may well choose to enhance these with
> many further facilities. 
> 
> 
>>This document should be of interest: http://www.nsa.gov/selinux/
> 
> 
> Interesting, but there's nothing new to me here. AdaOS will be designed to
> conform to the requirements of the TCSEC level B2 or B1. This exceeds what
> is offered by SE Linux (or any of the BSDs).
> 
> The security mechanisms of AdaOS will be built into each object (class),
> and will be tailored to the specific needs of that object (class); this is
> a considerably more sophisticated approach than any isomorphic OS can
> offer.
> 

To see a fairly complete implementation of this type of security system
see the latest versions of Sun Solaris.  You will find how it all works
under the heading "Role Based Access Control".  If you want the data 
labeling controls as well see Manditory Access Control under Sun's
Trusted Solaris Product.


-- 
Mark Biggar
mark.a.biggar@attbi.com

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[achrist]

Mark Biggar wrote:
> 
> 
> To see a fairly complete implementation of this type of security system
> see the latest versions of Sun Solaris.  You will find how it all works
> under the heading "Role Based Access Control".  If you want the data
> labeling controls as well see Manditory Access Control under Sun's
> Trusted Solaris Product.
> 

There is new healthcare privacy legislation in the US.  It mandates
the roll-based approach.  (For example, at the doctor's office, the
tech who draws blood doesn't get to see your financial info, the 
person who schedules appointments doesn't get to see your X-rays, 
the person who bills your insurer doesn't get to know your hospital
room number, etc, etc).  This approach doesn't relate well to simple
security by clearance levels, it's all based on what the user is 
doing.  If you  do two different jobs, you may have to use two
different sign-ons -- won't that be fun?  And unless your system 
has separate screens designed for each role, you'll likely have to 
control  visibility of the items on your screens, field-by-field and
role by role.  None of this is in any sense optional.

Healthcare being perhaps the only business bigger than national defense
in the US,  you can guess why Sun is going this way.   So, if AdaOS
has security based on clearance levels and not roles, it is 
irrelevant for meeting mandatory federal requirements for healthcare
organizations -- and that's perhaps the biggest industry in the US. 


Al

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Nick Roberts]

On Sun, 15 Sep 2002 01:38:25 -0700, achrist@easystreet.com strongly typed:

>...
>Healthcare being perhaps the only business bigger than national defense
>in the US,  you can guess why Sun is going this way.   So, if AdaOS
>has security based on clearance levels and not roles, it is 
>irrelevant for meeting mandatory federal requirements for healthcare
>organizations -- and that's perhaps the biggest industry in the US. 

As you will see from my other post, AdaOS will offer role-based security,
as well as mandatory (clearance levels).

-- 
Nick Roberts
Per Ardua ad Disastra

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Larry Kilgallen]

In article <3d84f49b.1270762508@news.cis.dfn.de>, nickroberts@blueyonder.co.uk (Nick Roberts) writes:
> On Sun, 15 Sep 2002 01:38:25 -0700, achrist@easystreet.com strongly typed:
> 
>>...
>>Healthcare being perhaps the only business bigger than national defense
>>in the US,  you can guess why Sun is going this way.   So, if AdaOS
>>has security based on clearance levels and not roles, it is 
>>irrelevant for meeting mandatory federal requirements for healthcare
>>organizations -- and that's perhaps the biggest industry in the US. 
> 
> As you will see from my other post, AdaOS will offer role-based security,
> as well as mandatory (clearance levels).

I presume that means for Confidentiality you plan to include the Bell and
LaPadula compartments as well as levels.  How about the Biba Integrity
Extensions ?  The code to add that is trivial, since it is the logical
inverse of of Bell and LaPadula.

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Nick Roberts]

On 16 Sep 2002 09:26:53 -0600, Kilgallen@SpamCop.net (Larry Kilgallen)
strongly typed:

>> As you will see from my other post, AdaOS will offer role-based security,
>> as well as mandatory (clearance levels).
>
>I presume that means for Confidentiality you plan to include the Bell and
>LaPadula compartments as well as levels.  How about the Biba Integrity
>Extensions ?  The code to add that is trivial, since it is the logical
>inverse of of Bell and LaPadula.

Yes indeed, as per the TCSEC. (Essentially this is about the addition of
special categories to the basic levels.) I may even go a step further, and
add different confinement rule options to the categories.

-- 
Nick Roberts
Per Ardua ad Disastra

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Larry Kilgallen]

In article <3d868ca3.16065380@news.cis.dfn.de>, nickroberts@blueyonder.co.uk (Nick Roberts) writes:
> On 16 Sep 2002 09:26:53 -0600, Kilgallen@SpamCop.net (Larry Kilgallen)
> strongly typed:
> 
>>> As you will see from my other post, AdaOS will offer role-based security,
>>> as well as mandatory (clearance levels).
>>
>>I presume that means for Confidentiality you plan to include the Bell and
>>LaPadula compartments as well as levels.  How about the Biba Integrity
>>Extensions ?  The code to add that is trivial, since it is the logical
>>inverse of of Bell and LaPadula.
> 
> Yes indeed, as per the TCSEC. (Essentially this is about the addition of
> special categories to the basic levels.) I may even go a step further, and
> add different confinement rule options to the categories.

My emphasis on the Biba extensions is that technically they are not
included in the TCSEC, given the US government emphasis on keeping
the data secret but not on keeping the data correct (at the time
the TCSEC was written anyway).

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Nick Roberts]

On 16 Sep 2002 22:07:19 -0600, Kilgallen@SpamCop.net (Larry Kilgallen)
strongly typed:

>My emphasis on the Biba extensions is that technically they are not
>included in the TCSEC, given the US government emphasis on keeping
>the data secret but not on keeping the data correct (at the time
>the TCSEC was written anyway).

Well, _I'm_ concerned about data integrity! ;-) I shall implement integrity
controls as sophisticated as the confidentiality controls.

-- 
Nick Roberts
Per Ardua ad Disastra

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Larry Kilgallen]

In article <3d80b566.992395741@news.cis.dfn.de>, nickroberts@blueyonder.co.uk (Nick Roberts) writes:

> Currently, I plan to write 'native' device drivers (in Ada) for AdaOS. I
> intend to go to the original hardware documentation wherever posisble, and
> write the drivers properly. I don't think it's unfair to suggest that most
> device drivers (even commercial ones) tend be of remarkably poor quality.
> 
> Nevertheless, it would be really handy to provide compatibility layers that
> would allow existing device drivers to be used. At some stage, maybe we
> will have a go at this.

Existing device drivers that follow what model ?

Certainly that is fundamental to the design of the "native" device
driver system.  Unless you design it from the start, for instance,
it would be difficult to deliver ASTs.

RE: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Robert C. Leif]

One advantage of the two XML rendering languages, XSL and SVG, is that
their output could be used as the input to the printer or display device
driver or the XML rendering could be converted to some standard graphics
form such as Postscript or the HP printer language. The disadvantage is
that the two languages have neither a parent-child or sibling
relationship. 
XSL including its FO rendering methods is available from Apache. It is
written in Java. Unfortunately, my reading of their newsgroup indicates
the virtues of Ada. SVG is available from Adobe. Unfortunately, SVG is
described by DTDs, which are at least well written DTDs.
The next version of SVG, "Scalable Vector Graphics (SVG) 1.1
Specification
W3C Candidate Recommendation 30 April 2002" Includes both an IDL and a
Java binding. I believe that the IDL could be machine translated using
the technology for the Ada CORBA bindings. The DTD is really a group of
modular DTDs, which need some sort of a header in order to be translated
by XMLSpy into schemas.
Bob Leif
-----Original Message-----
From: comp.lang.ada-admin@ada.eu.org
[mailto:comp.lang.ada-admin@ada.eu.org] On Behalf Of Larry Kilgallen
Sent: Thursday, September 12, 2002 10:23 AM
To: comp.lang.ada@ada.eu.org
Subject: Re: Advantage of XML based GUI? (was Re: Ada-inspired
OS/Language)

In article <3d80b566.992395741@news.cis.dfn.de>,
nickroberts@blueyonder.co.uk (Nick Roberts) writes:

> Currently, I plan to write 'native' device drivers (in Ada) for AdaOS.
I
> intend to go to the original hardware documentation wherever posisble,
and
> write the drivers properly. I don't think it's unfair to suggest that
most
> device drivers (even commercial ones) tend be of remarkably poor
quality.
> 
> Nevertheless, it would be really handy to provide compatibility layers
that
> would allow existing device drivers to be used. At some stage, maybe
we
> will have a go at this.

Existing device drivers that follow what model ?

Certainly that is fundamental to the design of the "native" device
driver system.  Unless you design it from the start, for instance,
it would be difficult to deliver ASTs.

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Preben Randhol]

On Thu, 12 Sep 2002 17:45:01 -0700, Robert C. Leif wrote:
> One advantage of the two XML rendering languages, XSL and SVG, is that
> their output could be used as the input to the printer or display device
> driver or the XML rendering could be converted to some standard graphics
> form such as Postscript or the HP printer language. The disadvantage is
> that the two languages have neither a parent-child or sibling
> relationship. 

What about pictures you scan that is not SVG graphics?

Preben

RE: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Robert C. Leif]

Although vectors are efficient, they do not preclude the use of bit maps
where appropriate.
Bob Leif
-----Original Message-----
From: comp.lang.ada-admin@ada.eu.org
[mailto:comp.lang.ada-admin@ada.eu.org] On Behalf Of Preben Randhol
Sent: Friday, September 13, 2002 1:10 AM
To: comp.lang.ada@ada.eu.org
Subject: Re: Advantage of XML based GUI? (was Re: Ada-inspired
OS/Language)

On Thu, 12 Sep 2002 17:45:01 -0700, Robert C. Leif wrote:
> One advantage of the two XML rendering languages, XSL and SVG, is that
> their output could be used as the input to the printer or display
device
> driver or the XML rendering could be converted to some standard
graphics
> form such as Postscript or the HP printer language. The disadvantage
is
> that the two languages have neither a parent-child or sibling
> relationship. 

What about pictures you scan that is not SVG graphics?

Preben

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Nick Roberts]

On 12 Sep 2002 11:23:11 -0600, Kilgallen@SpamCop.net (Larry Kilgallen)
strongly typed:

>In article <3d80b566.992395741@news.cis.dfn.de>, nickroberts@blueyonder.co.uk (Nick Roberts) writes:
>
>> Currently, I plan to write 'native' device drivers (in Ada) for AdaOS. I
>> intend to go to the original hardware documentation wherever posisble, and
>> write the drivers properly. I don't think it's unfair to suggest that most
>> device drivers (even commercial ones) tend be of remarkably poor quality.
>> 
>> Nevertheless, it would be really handy to provide compatibility layers that
>> would allow existing device drivers to be used. At some stage, maybe we
>> will have a go at this.
>
>Existing device drivers that follow what model ?

Various and many. Maybe Linux. Maybe Windows. Probably a good plan would be
to start with one that is relatively simple (but still works), and work
towards the more sophisticated ones.

>Certainly that is fundamental to the design of the "native" device
>driver system.  

A device driver may require: port I/O; memory mapped I/O; interrupt handler
installation (and communication); communication with higher levels of
software. What else? The design of AdaOS is very flexible. I suspect most
DD environments would fit into AdaOS with little difficulty.

>Unless you design it from the start, for instance,
>it would be difficult to deliver ASTs.

What is an 'AST' in this context, please?

-- 
Nick Roberts

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Robert A Duff]

nickroberts@blueyonder.co.uk (Nick Roberts) writes:

> What is an 'AST' in this context, please?

I think Larry is referring to "Asynchronous System Trap", which is a VMS
thing, sort of like what Unix calls a "signal" and what some others call
a "software interrupt".

- Bob

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[David C. Hoos, Sr.]

----- Original Message ----- 
From: "Nick Roberts" <nickroberts@blueyonder.co.uk>
Newsgroups: comp.lang.ada
To: <comp.lang.ada@ada.eu.org>
Sent: September 12, 2002 8:53 PM
Subject: Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)


<snip>
> What is an 'AST' in this context, please?
In VMSese and AST is an Asynchronous System Trap similar to a UNIX signal.

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Larry Kilgallen]

In article <3d81400c.1027894993@news.cis.dfn.de>, nickroberts@blueyonder.co.uk (Nick Roberts) writes:
> On 12 Sep 2002 11:23:11 -0600, Kilgallen@SpamCop.net (Larry Kilgallen)
> strongly typed:

>>Existing device drivers that follow what model ?
> 
> Various and many. Maybe Linux. Maybe Windows. Probably a good plan would be
> to start with one that is relatively simple (but still works), and work
> towards the more sophisticated ones.

I believe such an approach would lead to a "tacked on" result such
as Windows NT, 2000, etc., with GUI from one heritage tacked onto an
operating system kernal from another.

>>Certainly that is fundamental to the design of the "native" device
>>driver system.  
> 
> A device driver may require: port I/O; memory mapped I/O; interrupt handler
> installation (and communication); communication with higher levels of
> software. What else? The design of AdaOS is very flexible. I suspect most
> DD environments would fit into AdaOS with little difficulty.

A Port and Class driver architecture, allowing in the simplest case
multiple drivers to share a single SCSI bus and in more complex cases
multiple ports to access the same device on a remote controller. For
a sanity check, I would suggest desk checking any design against
contemporary Fibre Channel hardware specifications to ensure that
IO to a specified disk can be handled over multiple paths.
 
>>Unless you design it from the start, for instance,
>>it would be difficult to deliver ASTs.
> 
> What is an 'AST' in this context, please?

Execution of code provided by the user, in user process context and
user processor mode when an IO request completes, independent of the
normal code flow of the user program.

======================

Of course I can only mention considerations I know about, I am sure
there are others with experience in other environments (IBM Parallel
sysplex, etc.) who should be consulted to understand high end device
driver environments.

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Nick Roberts]

On 13 Sep 2002 11:00:18 -0600, Kilgallen@SpamCop.net (Larry Kilgallen)
strongly typed:

>In article <3d81400c.1027894993@news.cis.dfn.de>, nickroberts@blueyonder.co.uk (Nick Roberts) writes:
>> On 12 Sep 2002 11:23:11 -0600, Kilgallen@SpamCop.net (Larry Kilgallen)
>> strongly typed:
>
>>>Existing device drivers that follow what model ?
>> 
>> Various and many. Maybe Linux. Maybe Windows. Probably a good plan would be
>> to start with one that is relatively simple (but still works), and work
>> towards the more sophisticated ones.
>
>I believe such an approach would lead to a "tacked on" result such
>as Windows NT, 2000, etc., with GUI from one heritage tacked onto an
>operating system kernal from another.

Yes indeed, I quite agree. It would be 'handy' to be able to import foreign
device drivers, but certainly not ideal. Ideally, our device drivers would
all be written natively for AdaOS.

>>>Certainly that is fundamental to the design of the "native" device
>>>driver system.  
>> 
>> A device driver may require: port I/O; memory mapped I/O; interrupt handler
>> installation (and communication); communication with higher levels of
>> software. What else? The design of AdaOS is very flexible. I suspect most
>> DD environments would fit into AdaOS with little difficulty.
>
>A Port and Class driver architecture, allowing in the simplest case
>multiple drivers to share a single SCSI bus and in more complex cases
>multiple ports to access the same device on a remote controller. 

This sounds like the kind of situation where one needs to 'fake' hardware,
by facilitating virtual ports and interrupts. I'm sure this could be added
to Bachar (the AdaOS microkernel) very readily.

>For a sanity check, I would suggest desk checking any design against
>contemporary Fibre Channel hardware specifications to ensure that
>IO to a specified disk can be handled over multiple paths.

Okay.

>>>Unless you design it from the start, for instance,
>>>it would be difficult to deliver ASTs.
>> 
>> What is an 'AST' in this context, please?
>
>Execution of code provided by the user, in user process context and
>user processor mode when an IO request completes, independent of the
>normal code flow of the user program.

The Bachar microkernel provides exactly this capability in two ways: a
privileged process is permitted to convert a thread into an interrupt
service routine; a thread can 'generate an exception' in another thread
(within the same process). So "no problemo", I think!

>======================
>
>Of course I can only mention considerations I know about, I am sure
>there are others with experience in other environments (IBM Parallel
>sysplex, etc.) who should be consulted to understand high end device
>driver environments.

In any event, I have designed Bachar (and other AdaOS bits) to be extremely
future-proof, so that if it becomes necessary to add special functionality
(to accomodate special hardware, for example), it should be relatively
painless to do it.

-- 
Nick Roberts

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[chris.danx]

Nick Roberts wrote:

> Currently, I plan to write 'native' device drivers (in Ada) for AdaOS. I
> intend to go to the original hardware documentation wherever posisble, and
> write the drivers properly. I don't think it's unfair to suggest that most
> device drivers (even commercial ones) tend be of remarkably poor quality.
> 
> Nevertheless, it would be really handy to provide compatibility layers that
> would allow existing device drivers to be used. At some stage, maybe we
> will have a go at this.

What about UDI?

It's not the most Ada thing in the universe but it's all about reuse, so 
perhaps it's worth considering?

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Nick Roberts]

On Thu, 12 Sep 2002 21:36:46 +0100, "chris.danx"
<spamoff.danx@ntlworld.com> strongly typed:

>Nick Roberts wrote:
>
>> Currently, I plan to write 'native' device drivers (in Ada) for AdaOS. I
>> intend to go to the original hardware documentation wherever posisble, and
>> write the drivers properly. I don't think it's unfair to suggest that most
>> device drivers (even commercial ones) tend be of remarkably poor quality.
>> 
>> Nevertheless, it would be really handy to provide compatibility layers that
>> would allow existing device drivers to be used. At some stage, maybe we
>> will have a go at this.
>
>What about UDI?
>
>It's not the most Ada thing in the universe but it's all about reuse, so 
>perhaps it's worth considering?

I have studied UDI in detail. It is completely based on the C language, and
so is unusable for us in that form.

However, it might well be a very promising initiative for me (perhaps as
part of a wider collaboration) to create (and agree) an Ada binding for
UDI.

-- 
Nick Roberts

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Marin David Condic]

Give a lot of thought to the device driver model. You'd like it to be as
dynamic as possible. Ultimately, you'd like for the end user to plug in
a new device and just plain have it work - but that would require
support within the hardware and may not be possible in the PC
architecture. But whatever you can do to minimize the need for the user
to mess around with installing and selecting drivers is going to be a
plus for the OS.

MDC
-- 
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jast.mil/

Send Replies To: m c o n d i c @ a c m . o r g

    "I'd trade it all for just a little more"
        --  Charles Montgomery Burns, [4F10]
======================================================================

Nick Roberts wrote:

> 
> Currently, I plan to write 'native' device drivers (in Ada) for AdaOS. I
> intend to go to the original hardware documentation wherever posisble, and
> write the drivers properly. I don't think it's unfair to suggest that most
> device drivers (even commercial ones) tend be of remarkably poor quality.
> 
> Nevertheless, it would be really handy to provide compatibility layers that
> would allow existing device drivers to be used. At some stage, maybe we
> will have a go at this.
>

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Robert A Duff]

Marin David Condic <mcondic.auntie.spam@acm.org> writes:

> ...But whatever you can do to minimize the need for the user
> to mess around with installing and selecting drivers is going to be a
> plus for the OS.

For each driver installed, the user should be required to reboot the
machine 6 times.  While standing on one foot.

- Bob

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Larry Kilgallen]

In article <wcchegt29hu.fsf@shell01.TheWorld.com>, Robert A Duff <bobduff@shell01.TheWorld.com> writes:
> Marin David Condic <mcondic.auntie.spam@acm.org> writes:
> 
>> ...But whatever you can do to minimize the need for the user
>> to mess around with installing and selecting drivers is going to be a
>> plus for the OS.
> 
> For each driver installed, the user should be required to reboot the
> machine 6 times.  While standing on one foot.

No, while standing on 5 feet !

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Marin David Condic]

Larry Kilgallen <Kilgallen@SpamCop.net> wrote in message
news:lWN852sx0i7b@eisner.encompasserve.org...
> In article <wcchegt29hu.fsf@shell01.TheWorld.com>, Robert A Duff
<bobduff@shell01.TheWorld.com> writes:
> > Marin David Condic <mcondic.auntie.spam@acm.org> writes:
> >
> >> ...But whatever you can do to minimize the need for the user
> >> to mess around with installing and selecting drivers is going to be a
> >> plus for the OS.
> >
> > For each driver installed, the user should be required to reboot the
> > machine 6 times.  While standing on one foot.
>
> No, while standing on 5 feet !

Well, just so long as I don't have to slaughter a goat and leave its
entrails on the keyboard at moonrise, I think it will be an improvement.

MDC
--
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jast.mil/

Send Replies To: m c o n d i c @ a c m . o r g

    "I'd trade it all for just a little more"
        --  Charles Montgomery Burns, [4F10]
======================================================================

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Preben Randhol]

On Sat, 14 Sep 2002 08:49:29 -0400, Marin David Condic wrote:
> Larry Kilgallen <Kilgallen@SpamCop.net> wrote in message
> news:lWN852sx0i7b@eisner.encompasserve.org...
>> In article <wcchegt29hu.fsf@shell01.TheWorld.com>, Robert A Duff
><bobduff@shell01.TheWorld.com> writes:
>> > Marin David Condic <mcondic.auntie.spam@acm.org> writes:
>> >
>> > For each driver installed, the user should be required to reboot the
>> > machine 6 times.  While standing on one foot.
>>
>> No, while standing on 5 feet !
> 
> Well, just so long as I don't have to slaughter a goat and leave its
> entrails on the keyboard at moonrise, I think it will be an improvement.

*Detected mouse movement! Please stand by while rebooting to update
changes.*

Installing software on Windows is a pain in the neck and extremely time
consuming compared to Linux. Besides with Linux you learn something.

Preben

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Marin David Condic]

Well, software installation and driver installatin are two different things.
Ultimately, you really don't want to have the end user install drivers at
all. A user should be able to plug the device into the computer and it
should just work - no more complicated than installing a new lamp in your
livingroom. The problem with that is that you either a) have to have some
kind of common interface(s) specified so that the OS can talk to any device
of a specific class via an existing driver or b) have the new hardware
somehow load up the driver to the OS when it is plugged in. In either case,
you're talking about support from the hardware that may not exist in the
environment of interest. If you could get all the hardware guys to agree on
a standard and get the industry to say "Going forward, it shall be thus..."
the end user would have a lot easier time with it, but that's been
notoriously hard to achieve.

MDC
--
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jast.mil/

Send Replies To: m c o n d i c @ a c m . o r g

    "I'd trade it all for just a little more"
        --  Charles Montgomery Burns, [4F10]
======================================================================

Preben Randhol <randhol+abuse@pvv.org> wrote in message
news:slrnaob4ja.o7.randhol+abuse@kiuk0152.chembio.ntnu.no...
>
> *Detected mouse movement! Please stand by while rebooting to update
> changes.*
>
> Installing software on Windows is a pain in the neck and extremely time
> consuming compared to Linux. Besides with Linux you learn something.
>

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Jeffrey Carter]

Robert Duff:
 > For each driver installed, the user should be required to reboot the
 > machine 6 times.  While standing on one foot.

Come now. This is Ada. Surely we can do better than those hackers at
MicrobeSquash. 12 times. While standing on one head.

-- 
Jeff Carter
"I blow my nose on you."
Monty Python & the Holy Grail

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Mats Karlssohn]

Jeffrey Carter wrote:
> 
> Robert Duff:
>  > For each driver installed, the user should be required to reboot the
>  > machine 6 times.  While standing on one foot.
> 
> Come now. This is Ada. Surely we can do better than those hackers at
> MicrobeSquash. 12 times. While standing on one head.

While standing on Bills head maybe ?

-- 
Mats Karlssohn, developer                         mailto:mats@mida.se  
Mida Systemutveckling AB                          http://www.mida.se
Box 64, S-732 22 ARBOGA, SWEDEN
Phone: +46-(0)589-89808   Fax: +46-(0)589-89809

Re: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Warren W. Gay VE3WWG]

Nick Roberts wrote:
> On Thu, 12 Sep 2002 08:02:53 +0000 (UTC), Preben Randhol
> <randhol+news@pvv.org> strongly typed:
>>On Wed, 11 Sep 2002 19:48:28 GMT, Nick Roberts wrote:
...
>>I have started to read a bit on AdaOS and I have two questions. When it
>>comes to device drivers how are they going to be implemented. I mean one
>>of the problems Linux has is that one have to make all the device
>>drivers onself and somethimes this means reverse engineering. Would one
>>need to do the same for AdaOS or could on easily port the linux
>>drivers?
> 
> Currently, I plan to write 'native' device drivers (in Ada) for AdaOS. I
> intend to go to the original hardware documentation wherever posisble, and
> write the drivers properly. I don't think it's unfair to suggest that most
> device drivers (even commercial ones) tend be of remarkably poor quality.

Just be aware that the specs don't always tell you the whole story.
By looking at the code for existing drivers you can save yourself
a lot of grief by examining the issues that the driver author had
to contend with. You'll also see where he had to punt. You might
punt differently, but looking at existing driver code can be
extremely helpful. 8-)

One area that is particularly sensitive in this area is "probing"
at initialization time. A lot of PC hardware presents a
"no win" situation -- particularly older hardware.
-- 
Warren W. Gay VE3WWG
http://home.cogeco.ca/~ve3wwg

Re: Ada-inspired OS/Language

[Peter Hermann]

Marin David Condic <mcondic.auntie.spam@acm.org> wrote:
> ... would constitute "A Good Start".
                       "A g u s t a"           :-)

Re: Ada-inspired OS/Language

[Samuel Tardieu]

"Warren W. Gay VE3WWG" <ve3wwg@cogeco.ca> wrote in message news:<3D628304.3040506@cogeco.ca>...

> Now we've regressed with the palms etc., back to this no-protection model.

Are you referring to Palm Inc. handhelds? (if you aren't, let's blame
the language barrier and ignore the rest of this message :-).

If this is the case, this is unfair to Palm Inc. Their handheld
devices contain a chip which prevents memory writes outside of the
heap (only one application can be running and thus access the heap at
a given time, and the stack is located in the heap space). When an
application wants to write into a so-called database (for example
another application or users persistent data), the PalmOS system call
checks the address, directs the external chip not to intercept the
next write operations if the address is correct (previously locked
memory chunk), do the write then tells the chip to prevent further
write operations in non-heap areas.

They do not have true memory protection (which would be useless in
their case), but they do offer a working alternative. That is why
their devices almost never loose data, even when they run bogus
applications (which can cause a data-safe device reset).

  Sam
-- 
  Samuel Tardieu -- http://www.rfc1149.net/sam -- sam@rfc1149.net

RE: Advantage of XML based GUI? (was Re: Ada-inspired OS/Language)

[Robert C. Leif]

They are an external object that is imported and located on the screen.
XSL will handle gifs, tifs, etc. The real problem is making SVG and XSL
work together. However, either would be sufficient to make XForms work.
Bob Leif

-----Original Message-----
From: comp.lang.ada-admin@ada.eu.org
[mailto:comp.lang.ada-admin@ada.eu.org] On Behalf Of Robert C. Leif
Sent: Friday, September 13, 2002 10:44 AM
To: comp.lang.ada@ada.eu.org
Subject: RE: Advantage of XML based GUI? (was Re: Ada-inspired
OS/Language)

Although vectors are efficient, they do not preclude the use of bit maps
where appropriate.
Bob Leif
-----Original Message-----
From: comp.lang.ada-admin@ada.eu.org
[mailto:comp.lang.ada-admin@ada.eu.org] On Behalf Of Preben Randhol
Sent: Friday, September 13, 2002 1:10 AM
To: comp.lang.ada@ada.eu.org
Subject: Re: Advantage of XML based GUI? (was Re: Ada-inspired
OS/Language)

On Thu, 12 Sep 2002 17:45:01 -0700, Robert C. Leif wrote:
> One advantage of the two XML rendering languages, XSL and SVG, is that
> their output could be used as the input to the printer or display
device
> driver or the XML rendering could be converted to some standard
graphics
> form such as Postscript or the HP printer language. The disadvantage
is
> that the two languages have neither a parent-child or sibling
> relationship. 

What about pictures you scan that is not SVG graphics?

Preben