This is G o o g l e's cache of http://www.lp.se/ftp/mailinglists/FREE-VMS.1997-02.
G o o g l e's cache is the snapshot that we took of the page as we crawled the web.
The page may have changed since that time. Click here for the current page without highlighting.
Google is not affiliated with the authors of this page nor responsible for its content. |
| These search terms have been highlighted: | levitte | programming |
|
|
Archive-Date: Sat, 01 Feb 1997 04:09:49 +0100
Sender: owner-free-vms@lp.se
Date: Fri, 31 Jan 1997 22:03:06 -0500 (EST)
From: EVERHART@Arisia.GCE.Com
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Message-ID: <970131220306.63@Arisia.GCE.Com>
Subject: Re: VMSBACKUP 4.1 uploaded
Physical backup is actually not all that hard, since you need only some
few headers and so on, and need nothing that is file-specific.
Image backup would wind up ODS-2 specific...probably not what you
need.
My reader code is embedded in the stuff I put on the freeware CD
(look under "fast backup" or the like). However, the V3 docset bits
that give backup format are needed to do a writer; I used the bits
I needed and ignored the rest. It isn't rocket science...the CRC
polynomials were CCITT and Autodin as I recall, for example...but
is a bit of work.
One advantage to physical backup is that if you can figure how to present
the thing to your system as a disk, the native routines for disk can
pick off what you like and you don't need to mess with how to select
parts. While it copies all the stuff, even what isn't used, if you
grab in big chunks and compress them, you can in principle come up
with a save format that wastes relatively little space on the media
yet gets all the data.
One more thought: for newer tape drives, the CRC and XOR blocks and so
forth are less important than they used to be; the drives:
1. Have lots of ECC, and
2. Can't usually get beyond bad spots on the tape anyway.
A simpler checking system could wind up being just as good, as a practical
matter. But you need to save file attributes, which could grow more
important with time.
glenn
================================================================================
Archive-Date: Tue, 04 Feb 1997 22:20:48 +0100
Sender: owner-free-vms@lp.se
Message-ID: <c=US%a=_%p=msft%l=RED-70-MSG-970204212003Z-616@INET-02-IMC.microsoft.com>
From: Chris Kaler <ckaler@microsoft.com>
Reply-To: Free-VMS@lp.se
To: "'Free-VMS@lp.se'" <Free-VMS@lp.se>
Subject: RE: FW: Just a thought
Date: Tue, 4 Feb 1997 13:20:03 -0800
OK, let me try to answer this...
I've done this before, and I believe we were quite successful. The
summary is this: you don't need the kernel immediately, you need
people to go off and write code.
I would have people with access to VMS machines go and re-write
all of the applications desired using the VMS RTL. This means DIFF,
BACKUP, SORT, SEARCH, EDT, whatever. This is a great deal of
work. This gives several benefits: (1) we now have "free" applications
for FreeVMS. (2) We now have a large code base to test the VMS RTL
when it is built. (3) DECUS users get lots of sample code. This could
easily occupy 50 people for many long nights.
In *parallel*, a set of people go a write as much of the VMS RTL as
is possible in portable C, preferably on a UNIX box. This gives several
benefits as well: (1) All of the apps now work on UNIX, NT, and PCs.
(2) we now have a solid VMS user-level application base. (3) DECUS
users get lots of examples. This is a big job and has benefits even if
we never build a full FreeVMS system. This is also a huge job - remember
that RMS is part of this! Also, this includes the DCL parsing functions.
You don't need DCL yet... we can glue that in later.
Also in *parallel*, a set of folks decide on a Kernel, and start building
the foundations of a VMS personality. I would strongly recommend
doing it in such a way that you could also have other personalities at
the same time, but that is just my opinion. This is important. If you
don't build the right base, the system will be useless.
The last part is to glue the first two together with the 3rd. This is where
DCL and low-level system services come in. This is when the plan
comes together ;-)
You don't have to start at the bottom. You can, assuming this is a
non-commercial venture, draw on people like myself and Bill who
have done this exact experiment before. In fact, starting at both
sides is so much more productive. This is essentially the approach
we settled on, and it works!
As for the choice of kernel... well I don't know much about the 4.4BSD
kernel... I hacked UNIX long before that. But I would strongly recommend
using a state-of-the-art micro-kernel. It will save so much more time
and energy. As well, the UNIX kernel is designed to do UNIX well.
Micro-kernels are designed to be more obliging to other operating systems.
I would strongly suggest that someone take the reigns and *egads* try to
write a plan down on how to approach this. I would love to do it, but I
still feel "shackled" by my former employer - so I can only office advice
and consul. It shouldn't be too hard to create a list of all the important
applications and all of the RTL functions. We can decide on a priority,
and create a sign-up on a Web site. This way people could really start
banging on the first two steps. We should also agree on the approach
and put it in a FAQ so that people don't keep asking the same questions.
As well, someone needs to drive picking a kernel and working on an
architecture. I, not surprisingly, would take the Intrepid architecture and
make some refinements, and then drill-down.
Just my two cents... Christopher
-----Original Message-----
From: Todd C. Campbell [SMTP:toddc@net-link.net]
Sent: Friday, January 31, 1997 1:16 PM
To: Free-VMS@lp.se
Subject: Re: FW: Just a thought
Chris Kaler wrote:
>
> So this thread has been coming and going since this mailing list
> started. To be honest, I'm tired of it, so I am turning on my
> flame thrower ;-).
>
> It is totally insane to even try to write a kernel from scratch.
> First of all, there are very few people who can do it right, and
> I can probably count them on my fingers (and yes, I have a
> standard set of digits). Second, it takes a shitlaod of time.
> Any idea how long the NT kernel took? The Mach kernel?
> Third, building a kernel that supports VMS... well NT doesn't
> event do it all yet!
I do understand what you are trying to say. The problem, however, is that aside
from the VMSBACKUP that may or may not actually work on FreeVMS, I have seen little
work out of this group. There seems to be less and less orginization as the months go
on. If there is to be a kernel that is to be built upon, then which one is it, where
can it be downloaded from and what platforms will it run on? I have offered both
machines and T1 to DS3 access, I have offered newsgroups or what ever it will take to
get a little more orginized discussions, I have had no response.
Yes, I have heard about binary compatiblity with FreeBSD and VMS, but now that I have
been told there is no such animal, I really have nothing to say. It seems like many
people would like to contribute their time, but they do not have the experience that
some of the rest of the group may have. Don't slam these people in front of the list,
e-mail that person privately, help them. Otherwise this group will only become a place
where FreeVMS will exist in theory, and you will scare off those lurkers that
aren't quite sure what they would like or be able to help with, nothing more will come
of it if we continue to punce on those who speak up with an idea or opinion.(I sure hope
you don't think you have all the answers, I know I don't)
================================================================================
Archive-Date: Wed, 05 Feb 1997 00:30:40 +0100
Sender: owner-free-vms@lp.se
From: Kenn Humborg <kenn@avalon.wombat.ie>
Message-ID: <199702042330.XAA01208@avalon.wombat.ie>
Subject: Suitability of Java
To: Free-VMS@lp.se
Date: Tue, 4 Feb 1997 23:29:59 +0000 (GMT)
CC: kenn@avalon.wombat.ie (Kenn Humborg)
Reply-To: Free-VMS@lp.se
Content-Type: text
Hi,
I posted a while back about my intention to write a mail reader with
a VMSMail-style frontend. At the time, my intention was to use it as
an excuse to learn C++ and do an OO design in a language that actually
supports it well. However, over the last few days, I've started
looking into Java. It seems to be _much_ simpler than C++ and has the
advantage of portability to the platforms that I initially want to target
(namely Linux and Win95).
Does anybody have any thoughts on the possibility of porting the JavaVM
to VMS? I doubt I'd be capable of doing it myself (and besides I don't
have a VMS machine to play on), but I can't imagine in would be too
difficult.
Might this provide a way to ease the initial development of FreeVMS?
The non-performance-critical apps could be implemented in Java.
Just my 2p.
--
Kenn kenn@wombat.ie, kenn@iol.ie
"It would appear that we have reached the limits of what is possible to
achieve with computer technology . . . although such statements tend to
sound pretty silly in 5 years." --John Von Neuman (ca.1949)
================================================================================
Archive-Date: Wed, 05 Feb 1997 01:58:46 +0100
Sender: owner-free-vms@lp.se
MIME-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Date: Wed, 5 Feb 1997 01:58:41 MET
Message-ID: <13047.59729.271759.feedmail.pl5.LEVITTE@devil.bofh.se>
From: Richard Levitte - VMS Whacker <levitte@lp.se>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: RE: FW: Just a thought
From: Chris Kaler <ckaler@microsoft.com>
You don't have to start at the bottom. You can, assuming this is a
non-commercial venture, draw on people like myself and Bill who
^^^^^^^^^^^^^^^^^^^^^^
If I have any say in this project (and I damn well hope I have), that's
what it's going to be.
I would strongly suggest that someone take the reigns and *egads* try to
write a plan down on how to approach this. I would love to do it, but I
still feel "shackled" by my former employer - so I can only office advice
and consul.
I plan to do exactly that (writing at least a start of a plan, because
what I have written so far isn't even there in my opinion), just I get
out of this damn flue (I'm getting better, but not really there yet).
Chris, I will take on your offer as advisor, as I will throw the plans
I make and listen to your opinion. Yes, I will throw the plans on the
list as well...
--
R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma; SWEDEN
Tel: +46-8-26 52 47; Cel: +46-10-222 64 05; No fax right now
PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65
Check http://www.lp.se/~levitte for my public key. bastard@bofh.se
================================================================================
Archive-Date: Wed, 05 Feb 1997 02:12:51 +0100
Sender: owner-free-vms@lp.se
Message-ID: <c=US%a=_%p=msft%l=RED-70-MSG-970205011210Z-2074@INET-02-IMC.microsoft.com>
From: Chris Kaler <ckaler@microsoft.com>
Reply-To: Free-VMS@lp.se
To: "'Free-VMS@lp.se'" <Free-VMS@lp.se>
Subject: RE: Suitability of Java
Date: Tue, 4 Feb 1997 17:12:10 -0800
You should send mail to Sue Denham at VMS. Denham@star.zko.dec.com
She is the OpenVMS Internet Product Suite Product Manager. You should
see if VMS has any intention of doing this.
Chris
-----Original Message-----
From: Kenn Humborg [SMTP:kenn@avalon.wombat.ie]
Sent: Tuesday, February 04, 1997 3:30 PM
To: Free-VMS@lp.se
Cc: kenn@avalon.wombat.ie
Subject: Suitability of Java
Hi,
I posted a while back about my intention to write a mail reader with
a VMSMail-style frontend. At the time, my intention was to use it as
an excuse to learn C++ and do an OO design in a language that actually
supports it well. However, over the last few days, I've started
looking into Java. It seems to be _much_ simpler than C++ and has the
advantage of portability to the platforms that I initially want to target
(namely Linux and Win95).
Does anybody have any thoughts on the possibility of porting the JavaVM
to VMS? I doubt I'd be capable of doing it myself (and besides I don't
have a VMS machine to play on), but I can't imagine in would be too
difficult.
Might this provide a way to ease the initial development of FreeVMS?
The non-performance-critical apps could be implemented in Java.
Just my 2p.
--
Kenn kenn@wombat.ie, kenn@iol.ie
"It would appear that we have reached the limits of what is possible to
achieve with computer technology . . . although such statements tend to
sound pretty silly in 5 years." --John Von Neuman (ca.1949)
================================================================================
Archive-Date: Wed, 05 Feb 1997 11:24:53 +0100
Sender: owner-free-vms@lp.se
Date: Wed, 5 Feb 1997 5:24:40 -0500 (EST)
From: jhjennis2@mmm.com
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Message-ID: <970205052440.21e03bad@mdw>
Subject: RE: Suitability of Java
Kenn,
There is a public domain version of Java (guavac) that runs on
Linux (and some other Os's as well). You can subscribe to the
mailing list at Majordomo@eecs.com.
Some time back, Richard expressed an interest in looking at a
VMS port, but I do not know if he is still interested.
Regards,
Jim Jennis, Sr. Manufacturing Specialist
Imation Corp.
Printing & Publishing Systems
Manufacturing Systems/Networking/Internet Services
Middleway Plant
Middleway, WV. USA.
Internet: jhjennis2@mmm.com
================================================================================
Archive-Date: Wed, 05 Feb 1997 12:49:38 +0100
Sender: owner-free-vms@lp.se
MIME-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Date: Wed, 5 Feb 1997 12:49:35 MET
Message-ID: <13048.33247.271759.feedmail.pl5.LEVITTE@devil.bofh.se>
From: Richard Levitte - VMS Whacker <levitte@lp.se>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: RE: Suitability of Java
From: jhjennis2@mmm.com
There is a public domain version of Java (guavac) that runs on
Linux (and some other Os's as well). You can subscribe to the
mailing list at Majordomo@eecs.com.
Some time back, Richard expressed an interest in looking at a
VMS port, but I do not know if he is still interested.
I'm still interested, but I need to finish som important ports of GNU
programs first. Like autoconf.
--
R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma; SWEDEN
Tel: +46-8-26 52 47; Cel: +46-10-222 64 05; No fax right now
PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65
Check http://www.lp.se/~levitte for my public key. bastard@bofh.se
================================================================================
Archive-Date: Wed, 05 Feb 1997 13:14:52 +0100
Sender: owner-free-vms@lp.se
Message-ID: <c=SE%a=_%p=LB_Data_AB%l=BUTTER-970205121501Z-194@butter.lbdata.se>
From: Robert Nydahl <RNydahl@lbdata.se>
Reply-To: Free-VMS@lp.se
To: "'Free-VMS@lp.se'" <Free-VMS@lp.se>
Subject: RE: How to help on Free-VMS
Date: Wed, 5 Feb 1997 13:15:01 +0100
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
>>> From: Jim Kingdon <kingdonc@harvey.cyclic.com>
>>>
>>> Guest accounts on VMS systems are also probably available, although I
>>> don't want to put any words in the mouths of the people who might be
>>> willing to offer them--I'm sure they don't want to get inundated with
>>> requests.
>>
>>Such a system is being prepared. As a matter of fact, I should have
>>finished the preparations before christmas, but other events came in
>>my way.
Don't I recognize that from my own life? :-)
>>What is planned is really a double system: one computer you can log in
>>to, and that is "crash friendly", meaning that if someone screws up, he
>>or she will be forgiven, since this will be expected (after all, we'll
>>do quite a lot of experimenting, right? :-)). The other computer will
>>not be possible for anyone (well, almost) to log in to, but it will serve
>>one disk through VMS-to-VMS NFS (at least that's what's planned right now).
>>That machine will serve as FTP and HTTP server for the project, using
>>the data on the served disk. Or at least that's how I've understood the
>>plan (I'm currently just responsible of getting the "crash friendly"
>>system up and running). If I'm wrong, the owner of those computers will
>>probably correct me with a swift slapp :-).
>
No need to slap you as you are correct. How about trying to get things
going this
coming weekend? I'll, for once, will be home all weekend unless
something very
unusual shows up so you might as well just beep me, ok? Someone else
that
might be interested to help Richard and I running this system? I
believe that Richard
and I have a, hmmm, tendancy to be rather busy at times, overloaded with
work.
>/Robert Nydahl
================================================================================
Archive-Date: Wed, 05 Feb 1997 15:34:14 +0100
Sender: owner-free-vms@lp.se
Date: Wed, 05 Feb 1997 09:33:45 -0500 (EST)
From: "Andrew C. Stoffel (914) 574-4784" <acs@campus.com>
Reply-To: Free-VMS@lp.se
Subject: RE: Suitability of Java
To: Free-VMS@lp.se
Message-ID: <Pine.PMDF.3.91.970205092610.68217A-100000@RCCLNK.SUNYROCKLAND.EDU>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
On Wed, 5 Feb 1997 jhjennis2@mmm.com wrote:
[THis is sort of 'off-topic' for this list...]
> There is a public domain version of Java (guavac) that runs on
> Linux (and some other Os's as well). You can subscribe to the
> mailing list at Majordomo@eecs.com.
But a java virtual machine so we can RUN java 'programs' under VMS ?
I asked one of the maintainers of DEC's Internet Software CD (whatever
it's called?)(He listens in on the VMS Perl mailing list)
about its' availability & he indicated some thought was
going into it for a future Free Software CD release but nothing that
could be "put in writing" or promised...
If one of FreeVMS's goals s to be able to run VMS software doesn't
that give us Netscape ? Does anyone know if Netscape for VMS supports
java applets (meaning they have a java VM for VMS already ?)
> Some time back, Richard expressed an interest in looking at a
> VMS port, but I do not know if he is still interested.
Should we carry on a java+FreeVMS thread off-list ?
A preferred method of java for FreeVMS (let's do something
'value-added' :-)) for me would be to make it PART of FreeVMS in the
same sense Apple & other vendors are embedding java into their OS's...
-Andy-
------------------------------------------------------------------------
Andy Stoffel Project Consultant voice: (914) 574-4784
acs@campus.com http://acs.sunyrockland.edu/ fax: (914) 574-4354
Campus Consultants Group, Inc. A Campus America Company
[******* PGP public key: http://acs.sunyrockland.edu/pubkey.txt *******]
================================================================================
Archive-Date: Wed, 05 Feb 1997 18:27:32 +0100
Sender: owner-free-vms@lp.se
Date: Wed, 5 Feb 1997 12:27:45 -0500
Message-ID: <199702051727.MAA27874@harvey.cyclic.com>
From: Jim Kingdon <kingdonc@harvey.cyclic.com>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: Re: FW: Just a thought
> As a matter of fact, RMS usually asks contributors to the GNU project to
> sign a paper, and to have their boss sign a disclaimer saying they will
> not claim any rights to the program given to the FSF.
That is almost surely the safest way to go in terms of avoiding
questions about ownership down the road. However, in practice, my
experience is that it is often easier to rewrite code which people
send in rather than get that kind of paperwork.
What we have tried to do with CVS is to make sure that everyone who
sends in patches implicitly gives permission for their distribution
under the copying terms in question (they are directed to send them to
a designated submission address and all the documents which mention
that address also mention this notion). This is modeled after the
"submissions are our property" language that one sees in the context
of letters to the editor in places like newspapers.
I believe that linux, *BSD, etc., also have a more-or-less-informal
practice. It seems that in practice disputes are rare as long as the
copying terms are written and clear, that contributors understand that
those are the terms, and that people are willing to swiftly remove
code whose status is called into question (e.g. some versions of the
linux shadow password suite).
Free-VMS is nowhere near having anything resembling written copying
terms as described in the previous paragraph. The GPL has been
mentioned, and various people have complained about that (in favor of
a BSD-style license?), and the idea of of some kind of "non-commercial
use only" license has also been mentioned (there are a _lot_ of
different things that can mean, though).
Furthermore, it is not clear to me that Free-VMS is making any
progress towards agreeing on copying terms. It seems that people have
_widely_ differing assumptions.
================================================================================
Archive-Date: Wed, 05 Feb 1997 18:36:49 +0100
Sender: owner-free-vms@lp.se
MIME-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Date: Wed, 5 Feb 1997 18:36:45 MET
Message-ID: <13048.54077.271759.feedmail.pl5.LEVITTE@devil.bofh.se>
From: Richard Levitte - VMS Whacker <levitte@lp.se>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: Re: FW: Just a thought
From: Jim Kingdon <kingdonc@harvey.cyclic.com>
Furthermore, it is not clear to me that Free-VMS is making any
progress towards agreeing on copying terms. It seems that people have
_widely_ differing assumptions.
I agree with you. I am somewhat pro-GPL, but I can do with a BSD-style
license.
I might very well get tired of the endless tirade one day, and simply
assume the right to make a final decision, with the hope it will please
as many as possible :-).
Just have to get out of the flue...
--
R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma; SWEDEN
Tel: +46-8-26 52 47; Cel: +46-10-222 64 05; No fax right now
PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65
Check http://www.lp.se/~levitte for my public key. bastard@bofh.se
================================================================================
Archive-Date: Wed, 05 Feb 1997 18:43:52 +0100
Sender: owner-free-vms@lp.se
Date: Wed, 5 Feb 1997 12:44:02 -0500
Message-ID: <199702051744.MAA27965@harvey.cyclic.com>
From: Jim Kingdon <kingdonc@harvey.cyclic.com>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: Re: VMSBACKUP 4.1 uploaded
BCC:
> My reader code is embedded in the stuff I put on the freeware CD
I assume you are talking about the following from "OpenVMS Freeware (VS0185)":
[.FASTBACK]
Facility to access BACK/PHYS or HSC <Backup> saves as disks
And if I understand it correctly, this is only for physical backups, right?
Just want to make sure that I am correctly understanding the
implications (or non-implications) this has for VMSBACKUP.
================================================================================
Archive-Date: Wed, 05 Feb 1997 19:33:47 +0100
Sender: owner-free-vms@lp.se
Date: Wed, 5 Feb 1997 13:33:57 -0500
Message-ID: <199702051833.NAA28225@harvey.cyclic.com>
From: Jim Kingdon <kingdonc@harvey.cyclic.com>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: Re: FW: Just a thought
BCC:
> I might very well get tired of the endless tirade one day, and simply
> assume the right to make a final decision, with the hope it will please
> as many as possible :-).
Hmm. Having you assume the right to make various decisions might be
better than not having decisions get made.
But have you thought about setting up a core, like in NetBSD, FreeBSD,
XFree86, CVS, etc., to make decisions? The criteria used by these
groups boil down to more or less that one gets accepted into the core
by contributing to the project, and it is the existing members of the
core who judge whether the criteria have been met. We could start
with the people who have contributed something so far (you for the web
site and many things, Hunter Goatley for his FTP server, me for
VMSBACKUP, Kevin Handy for STR$ and BTRAN, maybe others if I'm
forgetting someone).
Of course the legitimacy of a core's power depends on the willingness
of the masses (particularly people who are or might be writing code,
but also users and even non-users to a lesser extent) to recognize
them as being worth listening to. I would assume it would be easier
to build this legitimacy for a group organized as I suggest than for
you to attempt to assume this role as an individual.
If this seems like a good idea, I can provide a lot more thoughts on
various details of how it might be set up.
================================================================================
Archive-Date: Wed, 05 Feb 1997 20:01:30 +0100
Sender: owner-free-vms@lp.se
Message-ID: <32F8E714.6212@net-link.net>
Date: Wed, 05 Feb 1997 14:01:24 -0600
From: "Todd C. Campbell" <toddc@net-link.net>
Reply-To: Free-VMS@lp.se
MIME-Version: 1.0
To: Free-VMS@lp.se
Subject: Re: FW: Just a thought
References: <199702051833.NAA28225@harvey.cyclic.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Jim Kingdon wrote:
>
> > I might very well get tired of the endless tirade one day, and simply
> > assume the right to make a final decision, with the hope it will please
> > as many as possible :-).
>
> Hmm. Having you assume the right to make various decisions might be
> better than not having decisions get made.
>
> But have you thought about setting up a core, like in NetBSD, FreeBSD,
> XFree86, CVS, etc., to make decisions? The criteria used by these
> groups boil down to more or less that one gets accepted into the core
> by contributing to the project, and it is the existing members of the
> core who judge whether the criteria have been met.
Yes, there should be some sort of core or stearing commitee formed to set standards,
making decisions based on what is discused in the group/focal groups, and to report
back to the group what has been decided on, etc.
oh and just an aside, what speeds/countries are we all coming in at/from?
================================================================================
Archive-Date: Wed, 05 Feb 1997 20:20:36 +0100
Sender: owner-free-vms@lp.se
MIME-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Date: Wed, 5 Feb 1997 20:20:31 MET
Message-ID: <13048.60303.271759.feedmail.pl5.LEVITTE@devil.bofh.se>
From: Richard Levitte - VMS Whacker <levitte@lp.se>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: Re: FW: Just a thought
From: "Todd C. Campbell" <toddc@net-link.net>
Yes, there should be some sort of core or stearing commitee formed to set
standards, making decisions based on what is discused in the group/focal
groups, and to report back to the group what has been decided on, etc.
Yup, that's my thought too, and has been since I mumbled the first
thoughts about nine months ago. Apparently, I didn't get that thought
out well enough. I had the hopes a core would form pretty much by
itself, with this list as starting point... Heh, seems like it might,
just not as fast as I had imagined.
oh and just an aside, what speeds/countries are we all coming in at/from?
Sweden. Speeds? *laugh* depends on which system I decide to work. I have
access to several systems I can play with, with network speeds in the 10 Mb
range. It's very possible that I will be able to play with computers
connected with fiber directly to the central swedish router for foreign
traffic. The computers I play with are VAXen, like 6300-200 and VS4000-60,
and Unix stations ranging from SparcStations running SunOS 4.1.3_U2 to
PC's running NetBSD (latest version a few months ago).
The computer running this list is a VS4000-60 sitting behind a 64kb line,
for the moment. It will pretty soon move to another network with higher
speed.
--
R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma; SWEDEN
Tel: +46-8-26 52 47; Cel: +46-10-222 64 05; No fax right now
PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65
Check http://www.lp.se/~levitte for my public key. bastard@bofh.se
================================================================================
Archive-Date: Wed, 05 Feb 1997 22:16:52 +0100
Sender: owner-free-vms@lp.se
Message-ID: <F14AAE60A252CF11B56100805F14DD0901D766A8@RED-70-MSG>
From: Chris Kaler <ckaler@microsoft.com>
Reply-To: Free-VMS@lp.se
To: "'Free-VMS@lp.se'" <Free-VMS@lp.se>
Subject: RE: Suitability of Java
Date: Wed, 5 Feb 1997 09:49:49 -0800
As of Fall 1996, the Netscape port for OpenVMS did not support
Java applets. I don't know the current status. You should ask
them directly. Denham@star.zko.dec.com -- she is the product
manager.
One of the problems with porting the VM is the stack size. The
original Sun code was very married to a 32-bit stack. This caused
problem getting it ported to Alpha (64-bits). I don't know if the
changes were folded back into the public software. If not, *you*
will face these problems as well.
Chris
-----Original Message-----
From: Andrew C. Stoffel (914) 574-4784
[SMTP:acs@campus.com]
Sent: Wednesday, February 05, 1997 6:34 AM
To: Free-VMS@lp.se
Subject: RE: Suitability of Java
On Wed, 5 Feb 1997 jhjennis2@mmm.com wrote:
[THis is sort of 'off-topic' for this list...]
> There is a public domain version of Java (guavac) that
runs on
> Linux (and some other Os's as well). You can subscribe
to the
> mailing list at Majordomo@eecs.com.
But a java virtual machine so we can RUN java 'programs'
under VMS ?
I asked one of the maintainers of DEC's Internet
Software CD (whatever
it's called?)(He listens in on the VMS Perl mailing
list)
about its' availability & he indicated some thought was
going into it for a future Free Software CD release but
nothing that
could be "put in writing" or promised...
If one of FreeVMS's goals s to be able to run VMS
software doesn't
that give us Netscape ? Does anyone know if Netscape for
VMS supports
java applets (meaning they have a java VM for VMS
already ?)
> Some time back, Richard expressed an interest in
looking at a
> VMS port, but I do not know if he is still interested.
Should we carry on a java+FreeVMS thread off-list ?
A preferred method of java for FreeVMS (let's do
something
'value-added' :-)) for me would be to make it PART of
FreeVMS in the
same sense Apple & other vendors are embedding java into
their OS's...
-Andy-
--------------------------------------------------------
----------------
Andy Stoffel Project Consultant
voice: (914) 574-4784
acs@campus.com http://acs.sunyrockland.edu/ fax:
(914) 574-4354
Campus Consultants Group, Inc. A Campus
America Company
[******* PGP public key:
http://acs.sunyrockland.edu/pubkey.txt *******]
================================================================================
Archive-Date: Thu, 06 Feb 1997 00:27:42 +0100
Sender: owner-free-vms@lp.se
Date: Wed, 5 Feb 1997 18:28:55 -0500 (EST)
From: EVERHART@Arisia.GCE.Com
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Message-ID: <970205182855.62@Arisia.GCE.Com>
Subject: Re: VMSBACKUP 4.1 uploaded
The utility can only make HSC backups, BACK/PHYSICAL backups, or images
of disks made with something like unix dd look like disks...but it can
do all those.
Not related to VMSBACKUP as such, but it does have some use as a backuo
p strategy.
glenn
================================================================================
Archive-Date: Thu, 06 Feb 1997 02:09:14 +0100
Sender: owner-free-vms@lp.se
MIME-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Date: Thu, 6 Feb 1997 02:09:10 MET
Message-ID: <13049.15686.271759.feedmail.pl5.LEVITTE@devil.bofh.se>
From: Richard Levitte - VMS Whacker <levitte@lp.se>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: RE: Suitability of Java
From: Chris Kaler <ckaler@microsoft.com>
As of Fall 1996, the Netscape port for OpenVMS did not support
Java applets. I don't know the current status. You should ask
them directly. Denham@star.zko.dec.com -- she is the product
manager.
The latest beta (3.0b4 released last week) does not yet do Java.
Not even on VAX.
--
R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma; SWEDEN
Tel: +46-8-26 52 47; Cel: +46-10-222 64 05; No fax right now
PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65
Check http://www.lp.se/~levitte for my public key. bastard@bofh.se
================================================================================
Archive-Date: Thu, 06 Feb 1997 08:54:56 +0100
Sender: owner-free-vms@lp.se
Date: Wed, 5 Feb 1997 23:54:46 -0800
From: Tony Konashenok <tonyk@sseos.lbl.gov>
Reply-To: Free-VMS@lp.se
Message-ID: <199702060754.XAA15471@sseos.lbl.gov>
To: free-vms@lp.se
Subject: Some fuel for discussion
Hello folks,
Not ashamed of being a self-appointed _advocatus diaboli_ on this list,
I would like to throw in a few highly dubious arguments for the sole purpose
of fueling up the discussion, and our collective thinking along with it.
1. Concerning intellectual property rights, we have already observed here
that even a crude product might be of potential interest to commercial
developers. Most such cases would not do any harm, but we should be aware
of the historical precedent and clearly delineate our property rights
*before* the code comes into public existence. There are a few different
variants of the legal framework, GPL/LGPL being the most widely used,
although certainly not the only one possible. So... why don't those of
us who...
...actually read and compared several licensing schemes...
...discussed them with lawyers...
...wrote a licensing scheme...
...know of some quirks of a particular license...
or whatever else seems relevant, actually let the rest of us know about
that, so that after a while we could simply vote for one of the existing
variants, or have a lawyer draw up something original?
COME ON, SPEAK UP!
2. Looking at other public projects, I dare say that we should probably
establish ourselves as a legal entity in the near future. This may seem
a little bit far-fetched, but we could (and probably should) think ahead
about the following: this is a public project, and it is striving to
create a system which will be freely distributable; this, however, does
not preclude us (or anyone else, for that matter) from doing *support* of
this system on a commercial basis. The proposed legal entity would take
control of this if we decide to do so, and it seems to be one of the
few non-sharklike ways of defraying those inevitable expenses of the
project, along with affording some kind of legal protection to the
team members. There are many more useful aspects of incorporation. Once
again, all arguments for and against it are cordially invited.
3. The discussion of underlying kernel seems to be alive again, and I have
a feeling that it will keep going on until someone actually writes something
at least slightly functional. On the other hand, if you remember, a few
months ago I raised the question about necessity of a stricter QA process
than for free unixes. So, in the same spirit, I suggest that all of us who
feel like it start without further deliberation to write code on *whatever*
substrate they might like, with subsequent evalualtion by the whole team
and the QA guys in particular. BTW, the very feeling of someone else doing
an alternative implementation will make us write better code! ...And I am
not excluding myself from the coders, either!
4. Chris mentioned the necessity of planning, and Richard stated his
intention to do it. Of course, it is hard to plan a project that is being
done on a strictly voluntary basis, so we should probably draw up our plans
in the form of a "roadmap", trying to enumerate the necessary points and
the many possible ways of going through them, rather than to suggest the
chronology and the sequence of goals. I do have one such roadmap in my
mind, and I will probably commit it to paper ...er, hard disk, in the coming
days, but it probably draws too much on my personal views of what the
system should look like, so if anyone has some general suggestions, now
is the time to yell them out.
Tony Konashenok
================================================================================
Archive-Date: Thu, 06 Feb 1997 15:35:22 +0100
Sender: owner-free-vms@lp.se
Subject: Re: FW: Just a thought
To: Free-VMS@lp.se
Date: Thu, 6 Feb 1997 09:34:35 -0500 (EST)
From: "christopher f. chiesa" <lvt-cfc@servtech.com>
Reply-To: Free-VMS@lp.se
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Message-ID: <32f9ebfb0.5f0f@cyber3.servtech.com>
Content-Transfer-Encoding: 7bit
Re: where we're all coming from...
Rochester, NY. Anywhere from 2400 to 14.4K from me to the ISP, and
God-only-knows (though it "feels" like medium-heavily-loaded thinwire
Ethernet, if that helps... ;-)) from the ISP outward...
It's nice to find a snippet of this discussion that I can understand,
relate to, and contribute to. I STILL haven't been able to make sense of
anything else you folks have been talking to, in the week-and-a-half or so
that I've suddenly started receiving traffic from this list... I'm a VMS
guru of eleven years' experience, but don't know diddly about Mach, Linux,
any kind of PC, and not a whole lot about Unix (just enough to get myself
into trouble, basically). I have very little idea just how I could
conceivably contribute to this project, particularly if the relatively
easy STR$ etc. stuff has already been done...
Oh well, I'll keep monitoring the list and maybe it will all suddenly
become clear someday. Either that or you folks will finish the product
and someday I'll acquire a system capable of even RUNNING FreeVMS.
(Ideally I'd want an Amiga version, but FAT CHANCE of that, I'm sure!)
Chris Chiesa
lvt-cfc@servtech.com
================================================================================
Archive-Date: Thu, 06 Feb 1997 15:56:22 +0100
Sender: owner-free-vms@lp.se
MIME-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Date: Thu, 6 Feb 1997 15:56:19 MET
Message-ID: <13049.65315.271759.feedmail.pl5.LEVITTE@devil.bofh.se>
From: Richard Levitte - VMS Whacker <levitte@lp.se>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: Re: FW: Just a thought
From: "christopher f. chiesa" <lvt-cfc@servtech.com>
(Ideally I'd want an Amiga version, but FAT CHANCE of that, I'm sure!)
If Mach can run on Amiga (I doubt it today, but you never know), it
shouldn't be impossible...
--
R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma; SWEDEN
Tel: +46-8-26 52 47; Cel: +46-10-222 64 05; No fax right now
PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65
Check http://www.lp.se/~levitte for my public key. bastard@bofh.se
================================================================================
Archive-Date: Thu, 06 Feb 1997 17:40:06 +0100
Sender: owner-free-vms@lp.se
Date: Thu, 06 Feb 1997 11:39:32 -0500 (EST)
From: "Andrew C. Stoffel (914) 574-4784" <acs@campus.com>
Reply-To: Free-VMS@lp.se
Subject: Re: FW: Just a thought
To: Free-VMS@lp.se
CC: ASTOFFEL@sunyrockland.edu
Message-ID: <Pine.PMDF.3.91.970206113416.68232A-100000@RCCLNK.SUNYROCKLAND.EDU>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
On Thu, 6 Feb 1997, christopher f. chiesa wrote:
> Re: where we're all coming from...
>
> Rochester, NY. Anywhere from 2400 to 14.4K from me to the ISP, and
That's not bad :-)... I'm in Suffern, NY (across the Hudson from NYC),
the company I work for is in Knoxville, TN, & I'm moving to southern
NH in approx. 1 year.....
> (Ideally I'd want an Amiga version, but FAT CHANCE of that, I'm sure!)
Well... since I personally want a PowerPC version (Mac hardware at
this point).... I plan on contributing in whatever ways I can to help
make FreeVMS be available for more than 1 CPU type... one reason I
would favor a Mach based approach....
-Andy-
------------------------------------------------------------------------
Andy Stoffel Project Consultant voice: (914) 574-4784
acs@campus.com http://acs.sunyrockland.edu/ fax: (914) 574-4354
Campus Consultants Group, Inc. A Campus America Company
[******* PGP public key: http://acs.sunyrockland.edu/pubkey.txt *******]
================================================================================
Archive-Date: Thu, 06 Feb 1997 20:53:41 +0100
Sender: owner-free-vms@lp.se
Subject: Re: FW: Just a thought
To: Free-VMS@lp.se
Date: Thu, 6 Feb 1997 14:53:06 -0500 (EST)
From: "christopher f. chiesa" <lvt-cfc@servtech.com>
Reply-To: Free-VMS@lp.se
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Message-ID: <32fa36a30.1eab@cyber3.servtech.com>
Content-Transfer-Encoding: 7bit
> From: "christopher f. chiesa" <lvt-cfc@servtech.com>
>
> (Ideally I'd want an Amiga version, but FAT CHANCE of that, I'm sure!)
>
> If Mach can run on Amiga (I doubt it today, but you never know), it
> shouldn't be impossible...
I have NO IDEA whether Mach can run on Amiga. I have yet to find out
anything whatsoever about Mach, other than its name and a vague impression
that it is a "realtime OS," perhaps such as might be used for "embedded"
system designs...
Chris Chiesa
lvt-cfc@servtech.com
================================================================================
Archive-Date: Thu, 06 Feb 1997 21:08:59 +0100
Sender: owner-free-vms@lp.se
Message-ID: <v03007804af1fe8452934@[140.186.88.220]>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Thu, 6 Feb 1997 15:00:39 -0500
To: Free-VMS@lp.se
From: Dick Munroe <munroe@acornsw.com>
Reply-To: Free-VMS@lp.se
Subject: Where we're from...
Massachusetts, Hudson formerly, now Belmont.
Many years with Digital (inside and out) starting in the late 60s pretty
much through today. Lots of VMS work (inside and out).
Check out my company's website for a brief rundown on what I/we do:
http://www.acornsw.com/
Dick
--
Dick Munroe munroe@acornsw.com
Acorn Software (508) 568 1618 x1
267 Cox St. FAX: 562 1133
Hudson, Ma. 01749 http://www.acornsw.com/
Need a web site? a web server? other web service? Contact us...
"They told me to get Windows 3.1 or better...so I bought a Macintosh"
"Then they told me to get Windows 95 or better...so I bought another Mac!"
"Now they tell me to get Window NT 4.0 Server or better...my new Mac is on
order!"
================================================================================
Archive-Date: Fri, 07 Feb 1997 04:07:45 +0100
Sender: owner-free-vms@lp.se
MIME-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Date: Fri, 7 Feb 1997 04:07:40 MET
Message-ID: <13050.43660.271759.feedmail.pl5.LEVITTE@devil.bofh.se>
From: Richard Levitte - VMS Whacker <levitte@lp.se>
Reply-To: Free-VMS@lp.se
To: free-vms@lp.se
Subject: I've made changes and additions to the WWW pages...
Among other things, the pages have now been move to it's own
domain: free-vms.lp.se. This because I wanted to have a higher
flexibility and ability to move them around, and secondly
because I'm about to start a few specialised mailinglists
under that domain.
I've changed and added to the WWW pages:
ftp://free-vms.lp.se/index.html (changed)
ftp://free-vms.lp.se/news.html (changed)
ftp://free-vms.lp.se/activity/index.html (changed)
ftp://free-vms.lp.se/activity/discussions.html (changed)
ftp://free-vms.lp.se/activity/discussions_old.html (added)
ftp://free-vms.lp.se/activity/names.html (changed)
ftp://free-vms.lp.se/activity/resources.html (changed)
ftp://free-vms.lp.se/activity/wishlist.html (changed)
ftp://free-vms.lp.se/activity/whatnow.html (added)
ftp://free-vms.lp.se/docs/index.html (changed)
Please check them out.
--
R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma; SWEDEN
Tel: +46-8-26 52 47; Cel: +46-10-222 64 05; No fax right now
PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65
Check http://www.lp.se/~levitte for my public key. bastard@bofh.se
================================================================================
Archive-Date: Fri, 07 Feb 1997 04:09:36 +0100
Sender: owner-free-vms@lp.se
MIME-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Date: Fri, 7 Feb 1997 04:09:33 MET
Message-ID: <13050.43773.271759.feedmail.pl5.LEVITTE@devil.bofh.se>
From: Richard Levitte - VMS Whacker <levitte@lp.se>
Reply-To: Free-VMS@lp.se
To: free-vms@lp.se
Subject: I've made changes and additions to the WWW pages...
It's worth mentioning that I'm somewhat tired at the moment being,
and that the work I've done today (which is half of what I intended
to do) might seem badly done. I'll appreciate pointers if you find
such errors.
--
R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma; SWEDEN
Tel: +46-8-26 52 47; Cel: +46-10-222 64 05; No fax right now
PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65
Check http://www.lp.se/~levitte for my public key. bastard@bofh.se
================================================================================
Archive-Date: Fri, 07 Feb 1997 04:28:49 +0100
Sender: owner-free-vms@lp.se
To: Free-VMS@lp.se
Subject: Re: I've made changes and additions to the WWW pages...
References: <13050.43660.271759.feedmail.pl5.LEVITTE@devil.bofh.se>
MIME-Version: 1.0 (generated by tm-edit 7.68)
Content-Type: text/plain; charset=US-ASCII
From: Assar Westerlund <assar@sics.se>
Reply-To: Free-VMS@lp.se
Date: 07 Feb 1997 04:28:34 +0100
Message-ID: <5l6805xpjw.fsf@assaris.sics.se>
Richard Levitte - VMS Whacker <levitte@lp.se> writes:
> I've changed and added to the WWW pages:
>
> ftp://free-vms.lp.se/index.html (changed)
> ftp://free-vms.lp.se/news.html (changed)
> ftp://free-vms.lp.se/activity/index.html (changed)
> ftp://free-vms.lp.se/activity/discussions.html (changed)
> ftp://free-vms.lp.se/activity/discussions_old.html (added)
> ftp://free-vms.lp.se/activity/names.html (changed)
> ftp://free-vms.lp.se/activity/resources.html (changed)
> ftp://free-vms.lp.se/activity/wishlist.html (changed)
> ftp://free-vms.lp.se/activity/whatnow.html (added)
> ftp://free-vms.lp.se/docs/index.html (changed)
s/ftp/http/g I guess?
================================================================================
Archive-Date: Fri, 07 Feb 1997 04:34:09 +0100
Sender: owner-free-vms@lp.se
MIME-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Date: Fri, 7 Feb 1997 04:34:03 MET
Message-ID: <13050.45243.271759.feedmail.pl5.LEVITTE@devil.bofh.se>
From: Richard Levitte - VMS Whacker <levitte@lp.se>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: Re: I've made changes and additions to the WWW pages...
From: Assar Westerlund <assar@sics.se>
s/ftp/http/g I guess?
You mean `s/ftp/http/' :-). You DO mean that I was supposed to use
EDIT/EDT, don't you?
Otherwise, you're perfectly correct. Told you I was tired...
--
R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma; SWEDEN
Tel: +46-8-26 52 47; Cel: +46-10-222 64 05; No fax right now
PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65
Check http://www.lp.se/~levitte for my public key. bastard@bofh.se
================================================================================
Archive-Date: Sat, 08 Feb 1997 02:15:13 +0100
Sender: owner-free-vms@lp.se
Date: Fri, 7 Feb 1997 17:14:58 -0800
From: Tony Konashenok <tonyk@sseos.lbl.gov>
Reply-To: Free-VMS@lp.se
Message-ID: <199702080114.RAA17992@sseos.lbl.gov>
To: free-vms@lp.se
Subject: A draft of the general layout
Hello folks:
Here are some crude notes on a possible design layout. Please comment on it!
------------
Starting theses:
1. Ring structure of the OS is used
2. Platform dependence should be confined to minimum amount of code
3. Parallel development of inner and outer rings should be allowed
4. API compatibility with VMS should be implemented.
From this, the following composition of rings is suggested:
Ring 0. The only ring that can see platform-dependent things.
I/O drivers
Interrupt management
Memory management at hardware level
Timer and clock services
Ring 1. Can see physical memory (all memory mapped in one big segment,
except maybe private structures of the ring 0).
Scheduler/dispatcher
Paging/swapping
Memory allocator and scavenger
File ACP's
Network and queue (i.e. non-file-structured) ACP's
Driver API
Interprocess communication
AST delivery
Collector of extraneous interrupts
Ring 2. Can use privileged system services. This is where the notion of
process and its memory space appears. Each process has its memory
space plus an appropriate chunk of system space for communications.
Task initiator (job controller)
Queue manager
Image activator(s)
Personality(ies) (as sets of nonprivileged system services)
Security manager
Connection listeners (these are "pipelines" between terminals/queues/
networks and shells)
Ring 3. The puny mortals.
Shell(s) (sitting on top of ring 2 personalities)
Utilities
Applications
Concerning parallel development: all code in ring 3 should be first
grafted onto real VMS -- to allow for early development and to ensure
API compatibility. On the other hand, all code in ring 0 can be replaced
with small stubs interfacing to the host operating system running on the
target platform (for example, to MS-DOS + DJGPP on a PC). Programming
interfaces of the ring 0 code should be discussed and fixed at early stages
of development, and should be possibly simple. If several OSes exist on
the target platform, the most primitive is used for hosting.
A scenario of a typical session.
An interrupt comes from somewhere, and interrupt manager of ring 0 handles
it. It was unexpected, so the interrupt is conveyed to the collector of
extraneous interrupts, which determines it might be a login request and
tells it to the job controller, which creates a process and, using the image
activator, starts a connection listener in its context. Listener invokes the
security manager to process a login, and upon success the image activator to
attach a personality and then a shell. The shell receives from the listener
the information typed by the user, and invokes appropriate applications
which are started by image activators.
Simplified design at early stages: nonpaged memory, no security, only
one user process allowed (and already initiated at boot time), fixed
personality, no network, primitive stub file-ACP (instead of calling
I/O driver calls the host system and uses host system's file structure).
We need the following input from the team:
1. Recommendations and changes to this model
2. Suggestions on programming interfaces between components.
3. Who wants to take which component, or has experience and can offer advice.
4. Who can write hardware-dependent things on which hardware.
5. What existing free source code can be used as a prototype for a component.
Tony Konashenok
================================================================================
Archive-Date: Sat, 08 Feb 1997 02:25:43 +0100
Sender: owner-free-vms@lp.se
MIME-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Date: Sat, 8 Feb 1997 02:25:38 MET
Message-ID: <13051.58402.271759.feedmail.pl5.LEVITTE@devil.bofh.se>
From: Richard Levitte - VMS Whacker <levitte@lp.se>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: Re: A draft of the general layout
From: Tony Konashenok <tonyk@sseos.lbl.gov>
Good start. Right now, I just have a few comments, and they all relate
to the fact that the plan is to build on top of Mach:
Ring 0. The only ring that can see platform-dependent things.
Is covered by Mach.
Ring 1. Can see physical memory (all memory mapped in one big segment,
except maybe private structures of the ring 0).
Is partly covered by Mach.
Ring 2. Can use privileged system services. This is where the notion of
process and its memory space appears. Each process has its memory
space plus an appropriate chunk of system space for communications.
I don't think Mach covers this, but I haven't got the docs easely reachable.
Ring 3. The puny mortals.
Definitelly not covered by Mach.
--
R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma; SWEDEN
Tel: +46-8-26 52 47; Cel: +46-10-222 64 05; No fax right now
PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65
Check http://www.lp.se/~levitte for my public key. bastard@bofh.se
================================================================================
Archive-Date: Sat, 08 Feb 1997 17:16:20 +0100
Sender: owner-free-vms@lp.se
Message-ID: <01BC15DB.963C3300@PRO4>
From: "John F. Roden" <j.roden@procyon.ie>
Reply-To: Free-VMS@lp.se
To: "'Free-VMS@lp.se'" <Free-VMS@lp.se>
Subject: Joining the HURD...
Date: Sat, 8 Feb 1997 16:16:31 -0000
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
I've been reading up a bit on the Mach kernel and taking a look at the HURD
(which is built on Mach).
I'd like to float a (late night) idea that might give FreeVMS a lot of flexibility...
The HURD designers have laid heavy emphasis on keeping as much of the
operating system as possible out of the kernel with the aim of allowing components
to be replaced easily by others. This makes it easy to test new elements
or to add new features.
The file system on the HURD is especially flexible. Each file has an associated set of
handlers which provide the various functions needed by file users (e.g create, open,
read, write, delete, etc). The standard handlers can be replaced for any file individually.
This allows anything to be made to look like a file. It will have a file name and an entry in
a directory but its data may not be on a disk. (Their example shows how to use FTP to
make remote files look as if they were part of the local file system.)
This arrangement is object-oriented: the files are objects with methods (create, open,
etc). The objects have names in a hierarchy (the file system). The object class is changed
by changing the handlers.
My thought is that the HURD approach could be extended to handle nearly all the components
in the O/S. The objects would have different sets of methods but all could exist in the
same name space. For instance a serial port might be called:
node.devices.serial-ports.txa0
where the dots separate the layers of the hierarchy. If VMS logical names were made to
apply to these names it would make component substitution possible for each user. E.g
the device above could be defined as TTA0:.
define/object/sep="." TTA0: node.devices.serial-ports.txa0
Every object would have to support a query method which would allow any caller to discover
its type. This would lead to a simpler DCL where SHOW would not need to be qualified
(e.g SHOW sys$print). All objects would also be protected with standard VMS mechanisms.
The methods of most objects could be made available to remote callers via RPC with the
result that the system would become a network O/S (e.g SHOW remote-node.sys$print)
This whole scheme would be implemented between the Mach kernel and the VMS system
services layer. The naming of the objects could be more general than the current
VMS file name convention (allow any characters, Unicode, etc) but it would be possible
to map VMS file names easily in the system service layer. This would allow other personalities
for FreeVMS.
I have the impression that a lot of the thinking needed to take this route has already been done
with the GNU HURD O/S. These ideas have a parallel in OLE-2 and COM.
I'll be exploring this a bit further if I can get more info about the HURD... I know this is not at
the practical end of the scale but it might be of interest.
John Roden
j.roden@procyon.ie
================================================================================
Archive-Date: Sat, 08 Feb 1997 22:46:31 +0100
Sender: owner-free-vms@lp.se
From: "Allen Family" <ballen@mho.net>
Reply-To: Free-VMS@lp.se
To: <Free-VMS@lp.se>
CC: <ballen@mho.net>
Subject: Kernels and VMS
Date: Sat, 8 Feb 1997 14:49:07 -0700
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Message-ID: <19970208214857232.AAA174@claude.mho.net>
I have been reading the discussions on kernels for Free-VMS. This is my
2 cents.
Any operating system today lives or dies based on the type of hardware it
can
access. A perfect example is OS/2. Many people would have switched to OS/2
if it could access their favorite hardware. This requires drivers. IBM
could not
provide all the drivers nor could they get hardware manufactures to provide
the drivers either. Therefore, go to where the drivers are. We must pick a
kernel that has an installed base of device drivers. The two biggest
operating
systems in existance today that have many device drivers are Linux and...
Windows/NT! remember, IBM actually had a version of OS/2 which worked
with MS Windows 3.1. This was primarily to get access to microsofts
installed driver base.
Linux supports many devices. I bought the Info Magic Linux 6 CD set. It
runs
on Alpha and Intel that I know of. Many people are adding to Linux every
day.
Linux will solve the driver problem if we make Free-VMS compatable with
Linux drivers. Of course, this will make the internals of Free-VMS
different
from "real VMS". Linux does not have an asynchronous I/O subsystem like
"real VMS". Linux does not have ASTs like "real VMS". How much pain will
be envolved in adding these to Linux to build Free-VMS? This is a cheap
development platform. Being a VMS bigot I love the idea of using the Linux
kernel as
a starting point since this would result in Free-VMS assimilating Linux (
yikes the Borg are here! ) which could cause many UNIX/Linux types to have
coranaries over Linux being polluted with VMS features like a real file
system, logical names and DCL. (Please no flames this is just good natured
fun. The UNIX guys are ahead of
us VMS guys in getting a cheap version of their faviorite operating system.
)
Windows/NT supports many devices too. I have read the Helen Custer book
on Windows/NT and see many similarities between WNT and VMS. WNT has
an asynch I/O subsystem like VMS and WNT supports APCs which are close
to ASTs. WNT gives us a cheap ( not as cheap as Linux) development
platform.
WNT supports the concept of sub-systems. Free-VMS could be written as a
WNT sub-system. I talked to a WNT developer the other day and he explained
that Microsoft requires a non-discloser statement and they will release the
kernel source and etc needed to develop a sub-system. The down side is that
WNT only runs on Alpha and Intel now. MIPS and Motorola have stopped
porting WNT to their platforms last I heard. Free-VMS would be tied to the
fate of Microsoft. This would force Free-VMS to be only 32bit until WNT
goes
64bit. WNT also supports installable file systems. To date it supports FAT,
NTFS and HPFS. This should make it easy to add Files-11.
Intel is the only microprocessor I know of that supports four hardware
execution
modes in addition to the VAX and Alpha. Linux and WNT both support a two
mode
architecture so they can execute on any modern microprocessor. The Motorola
and MIPS processors only support two modes as well.
I think either direction will give Free-VMS compatability with "real VMS"
at the source
level. The advantages and disadvantages are as follows:
Linux Advanatages
----------------------------------------------------------------------------
--------------------------------
1. Free-VMS will execute on any hardware platform it's ported to.
2. Free-VMS will be available on 32 and 64 bit platforms.
3. Free-VMS will have a large device driver base to choose from.
Linux Disadvantages
----------------------------------------------------------------------------
---------------------------------
1. The Linux kernel will have to be hacked to include ASTs among other
things.
2. Development tools not as powerful as WNTs.
WNT Advantages
----------------------------------------------------------------------------
---------------------------------
1. Free-VMS will have a large device driver base to choose from.
2. Better development environment.
3. No kernel hacking.
4. Installable file systems.
WNT Disadvantages
----------------------------------------------------------------------------
-----------------------------------
1. Tied to Microsoft.
2. 32 bit for now and 64 bit when WNT goes 64 bit.
I think both directions should be taken. The source should be the same for
both
projects and would result in a way to port "real VMS" code to WNT and give
us
VMS guys a comfortable home under the WNT flag as it were. The Linux
assimilation
would eventually give Free-VMS a broader hardware base.
Bruce Allen
ballen@mho.net
The above statements are not my own.
They all belong to Rush Limbaugh -- so sue him!
I'm just a mind numbed robot.
================================================================================
Archive-Date: Sun, 09 Feb 1997 00:36:39 +0100
Sender: owner-free-vms@lp.se
Date: Sat, 08 Feb 1997 18:35:27 EST
From: "Brian Schenkenberger, VAXman-" <system@TMESIS.COM>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Message-ID: <009AF977.511B18C9.4@TMESIS.COM>
Subject: RE: Kernels and VMS
"Allen Family" <ballen@mho.net> sez:
Ok. Yes, I'm here on this list! I've been here on this list for months
and quiet and only joined out of curiosity. It's mostly been entertain-
ment; however, this guys posting is most disturbing.
>I have been reading the discussions on kernels for Free-VMS. This is my
>2 cents.
>{... really big
>WNT Advantages
>----------------------------------------------------------------------------
>---------------------------------
>1. Free-VMS will have a large device driver base to choose from.
>2. Better development environment.
>3. No kernel hacking.
>4. Installable file systems.
5. Poor, poor Mr. Gates makes more $$$.
6. Poor, poor Micro$aps make more $$$.
7. Micro$aps get to impede the development and future of real computing
by collecting more $$$ for their rediculous contributions to myriad
weekly-contrived/weekly-survived "industry" standards.
8. Free-VMS ain't free no more but puts more money into poor, poor Mr.
Gates pockets.
9. Bill Gates gets to be "pedestalled" for having the genius to create
yet another operating system when he clearly hasn't the wherewithall
to program his way out of a paper bag.
A. Higher stock prices for Micor$aps shares driven higher by the fervor
of a bunch of money grubbing wall-street morons that wouldn't know a
computer from a toaster.
Would you like yours light or dark?
[ (LIGHT) ] [ (DARK) ]
With Butter or Margarine?
[ (BUTTER) ] [ (OLEO) ]
General protection fault in crumb catcher.
Abort, Retry, Fail?
>WNT Disadvantages
>----------------------------------------------------------------------------
>-----------------------------------
>1. Tied to Microsoft.
>2. 32 bit for now and 64 bit when WNT goes 64 bit.
3. Sucks $$$ out of your pocket and places it into poor, poor Mr. Gates'
pockets. Could possibly be a tax write-off as a relocation expense.
4. One word: MicroSoft
5. ...
Sorry, I'd write more but the sobbing over the condition of poor, poor
Mr. Gates has short circuited my keyboard and made viewing the monitor
too blurry to continue...
>I think both directions should be taken. The source should be the same for
>both
>projects and would result in a way to port "real VMS" code to WNT and give
>us
>VMS guys a comfortable home under the WNT flag as it were. The Linux
>assimilation
>would eventually give Free-VMS a broader hardware base.
Ok. All funnin' aside:
This Free-VMS project seems to have no beginning and no end. Sounds like
the tennents of judao-christian theology more than an effort to work out
a viable alternative to paying for VMS. However, if you're going to lay
out $$$ for WNT to run on your favorite piece of hardware, why not lay a
few more down on Digital and purchase a superlative piece of engineering
effort in both hardware (Alpha)and software (VMS)? I fail to see how a
Free-VMS is free if your toading your hard earned money off to poor, poor
Mr. Gates. Buy an Alpha running VMS and produce an extensive collection
of free apps to run on the Alpha.
Also, the kernel, IMHO, is VMS. If you want to run DCL on a PC, look at
PC-DCL!
For my money, an Alpha and a VMS license ain't all that costly.
>I'm just a mind numbed robot.
Overdosing on Micro$aps, no doubt.
My $.02 worth to help finance the Free-VMS project! :) Does this mean
it's no longer free? :)
VAXman- VAXman@TMESIS.COM
================================================================================
Archive-Date: Sun, 09 Feb 1997 02:49:39 +0100
Sender: owner-free-vms@lp.se
From: "Allen Family" <ballen@mho.net>
Reply-To: Free-VMS@lp.se
To: <Free-VMS@lp.se>
CC: <ballen@mho.net>
Subject: RE: Kernels and VMS
Date: Sat, 8 Feb 1997 18:52:22 -0700
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Message-ID: <19970209015210386.AAA157@claude.mho.net>
Hey VAXman-
Since you are replying to my email pal I guess we can assume you
don't have 7 Billion in your pocket either do you ( Bill had 14 until
he got married now it's community property ).
So where is your software empire? I'm not out to make Microsoft
rich. They already are. I happen to admire Mr. Gates. The funny thing
is that if Digital had gotten a CLUE back in the early 80s VMS
might be on every desktop today and UNIX and Microsoft would
have been just another bad dream. Buy a DEC Rainbow and buy your
formatted floppies from DEC. Who was running the company then Homer
Simpson? DEC wants everyone to join DECUS and give their homemade
software away for FREE. Sounds a lot like Microsoft to me.
As I have seen in most software advertisements such as Borlands
Delphi etc. The DEVELOPER is paid for the product not
Microsoft. If Free-VMS did run as a sub-system of WNT then the
owner of it could either sell it or give it way.
OK. All fun'n aside.
You can't get a better OS than OpenVMS. Free-VMS could become
a movement to get Digital to lower the price on OpenVMS for the
rest of us.
Bruce Allen
ballen@mho.net
The above opinions are not my own.
They all belong to Rush Limbaugh --so sue him.
I'm just a minded numbed (Microsoft?) robot.
================================================================================
Archive-Date: Sun, 09 Feb 1997 13:03:48 +0100
Sender: owner-free-vms@lp.se
Date: Sun, 09 Feb 1997 07:02:38 EST
From: "Brian Schenkenberger, VAXman-" <system@TMESIS.COM>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Message-ID: <009AF9DF.B2DAE275.3@TMESIS.COM>
Subject: RE: Kernels and VMS
>Hey VAXman-
Yes.
>Since you are replying to my email pal I guess we can assume you
>don't have 7 Billion in your pocket either do you ( Bill had 14 until
>he got married now it's community property ).
I had $14.00 im my pocket. Now my atty's got it all! :(
You admire Mr. Gates for his $$$. I've heard him speak about the
industry on numerous occasions and thay all sounded like an adver-
tisement for Micro$aps more than anything of an intellectual value.
>formatted floppies from DEC. Who was running the company then Homer
>Simpson? DEC wants everyone to join DECUS and give their homemade
>software away for FREE. Sounds a lot like Microsoft to me.
Funny you should say that. I still have a news article with K.O.
showing off a VAX9009 MCUs. I have often thought he looked like
Homer Simpson in that picture. :)
>OK. All fun'n aside.
>
>You can't get a better OS than OpenVMS.
!
> Free-VMS could become
>a movement to get Digital to lower the price on OpenVMS for the
>rest of us.
and if a whole group starts singing, "You can get anything you want..."
they'll think its a movement. (Arlo) I'd hope I could take out Free-
VMS and play with it more than just once a year on Thanksgiving day! :)
I do find the price steps between an Alpha running:
VMS = $$$
UNIX = $$
WNT = $
to be a bit self-defeating for Digital. OTOH, I do not think VMS is
for the masses. Don't get me wrong, I'd love to see the PC packin'
world come to the realization that there is more to computing than a
word processor and Doom. Not when I walk into the local Barnes and
Nobel and see bookrack after bookrack with 1 1/2" thick volume of
books with titles like "DOS for Dimwits". If the general public is
that computer forlorn, they'd need a Library of Congress sized "VMS
is for Dimwits" just to do the basic.
This thread is NOT an attack on the Free-VMS project. No, not at all.
I'd actually like to see it to fuition. I do stronly disagree with
anything that ties it in anyway to Micro$aps. Then the Free in Free-
VMS will become moot and silent -- like Open in OpenVMS!
VAXman- VAXman@TMESIS.COM
================================================================================
Archive-Date: Sun, 09 Feb 1997 22:33:01 +0100
Sender: owner-free-vms@lp.se
Date: Sun, 9 Feb 1997 16:33:16 -0500
Message-ID: <199702092133.QAA28688@harvey.cyclic.com>
From: Jim Kingdon <kingdonc@harvey.cyclic.com>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: Re: FW: Just a thought
> I have very little idea just how I could conceivably contribute to
> this project, particularly if the relatively easy STR$ etc. stuff has
> already been done...
Last I looked, the STR$ routines had merely been started, not
finished. I would assume that help would be welcome. Also, there are
other parts of the system which are similarly easy. Browsing through
my manual I see LIB$MOVTC (straightforward), LIB$SCOPY_DXDX (more
string type stuff), LIB$SYS_ASCTIM (a thin layer on top of SYS$ASCTIM,
unless I'm missing something), LIB$DATE_TIME (not hard to implement in
terms of SYS$GETTIM and SYS$ASCTIM), SYS$FAO (someone said they were
going to work on it, but I don't think we've heard anything). SYS$FAO
is larger than the other ones I mention but I don't think any piece of
it is particularly tricky--it is easier than printf, for which you can
find free source code, because it lacks floating point. There are
plenty more examples where those came from...
If none of these sound doable, feel free to let us know what you'd
like instead. There is so much to do, that chances are we can find
projects for a variety of levels of skills and interests.
================================================================================
Archive-Date: Sun, 09 Feb 1997 22:47:43 +0100
Sender: owner-free-vms@lp.se
Date: Sun, 9 Feb 1997 16:48:51 -0500 (EST)
From: EVERHART@Arisia.GCE.Com
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Message-ID: <970209164851.e2@Arisia.GCE.Com>
Subject: RE: Kernels and VMS
Re:
>DEC wants everyone to join DECUS and give their homemade
>software away for FREE. Sounds a lot like Microsoft to me.
In fact, DECUS has been around quite a long time, and has distributed
software for years that was either essential for serious users of some
DEC operating systems or which competed against DEC products or both
as well as lots of other material. Fact is, those of us who have
participated have done so because we've found that we benefit from
the use of others' examples, and want to help others benefit from
outs...kind of a "what goes around comes around" thing.
DECUS is not about making money for anyone. It's about people sharing
expertise and information. Has nothing to do with commercial motives,
save perhaps very indirectly in that knowledgeable people are likely to
develop better software than people who are not and we all gain by
mutual aid.
There are still some folks around DEC that remember and value that
kind of attitude. Once it was pretty well the universally held attitude
among computerists. It may not be that way now, but at least try to
understand what it is correctly.
You might care to ask too what benefit there is to what you do as a software
developer. Writing something you hope will make you rich is fine. But
also, writing something whose sharing will contribute to the sum of human
knowing is at least an activity you can look back on and know that
your contribution will have had its benefit as long as civilization exists
to build on it, in however small a way. There are occupations which
have a lot harder stretch to convince their followers of any such thing.
Glenn C. Everhart
DECUS VMS SIG Tapecopy Coordinator
================================================================================
Archive-Date: Mon, 10 Feb 1997 12:59:34 +0100
Sender: owner-free-vms@lp.se
From: Gary.Gale@mail.sema.co.uk
Reply-To: Free-VMS@lp.se
Message-ID: <199702101202.MAA05226@gateway1.sema.co.uk>
To: Free-VMS@lp.se
Subject: A Potential Boost for Free-VMS
Date: 10 Feb 97 11:40:00 GMT
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Hey all,
The discussions re to Mach or not to Mach and to Hurd and not to Hurd
seem to be getting a tad frantic - so I thought I'd interject with
something which _may_ give the whole project a boost (and may not - we'll
have to wait and see).
A colleague at the company I'm currently contracting for has just joined
us after a number of years with Digital here in the UK. The topic of
Free-VMS came up and I was informed that
"Digital has released about 60% of VMS as shareware"
Now from what I have managed to glean, this mythical 60% included both
kernel subsystems and user mode applications - in other words a sizeable
chunk of DCL and the innards of VMS itself.
I'm awaiting further information prior to passing it on to the group -
but and this is a big BUT (3 miles high, neon lights, you get the idea)
this could be just the boost needed to kick start some serious
development.
Watch this space ...
Gary Gale
SMS Unix & Gateways Team
Sema Group Telecoms, Reading, United Kingdom
+44 (0) 1734 64 1686 +44 (0) 973 188380
gary.gale@mail.sema.co.uk / gary.g@dial.pipex.com / 100742,400
================================================================================
Archive-Date: Mon, 10 Feb 1997 16:28:26 +0100
Sender: owner-free-vms@lp.se
Date: Mon, 10 Feb 1997 10:22:43 -0400 (EDT)
From: "Like a tea tray in the sky..." <tom.otoole@jhu.edu>
Reply-To: Free-VMS@lp.se
Subject: RE: Kernels and VMS
To: Free-VMS@lp.se
Message-ID: <01IF8XCDOC048WW40S@JHUVMS.HCF.JHU.EDU>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
>Hey VAXman-
>[blah blah standard 'how come you not as rich as bill gates' claptrap removed]
>So where is your software empire? I'm not out to make Microsoft
>rich. They already are.
Maybe VAXman has a shred of business ethics and morality...
>I happen to admire Mr. Gates.
I guess I could understand that if greed, underhandedness, double-crossing
and anti-competitive trade practices are worthy of admiration.
>They all belong to Rush Limbaugh --so sue him.
Q:What did bill gates say to the hotdog vendor
A: Make me the one with everything.
-Tom O'Toole
================================================================================
Archive-Date: Mon, 10 Feb 1997 17:37:32 +0100
Sender: owner-free-vms@lp.se
Date: Mon, 10 Feb 1997 11:36:23 EST
From: "Brian Schenkenberger, VAXman-" <system@TMESIS.COM>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Message-ID: <009AFACF.1B4680A9.1@TMESIS.COM>
Subject: RE: Kernels and VMS
>>Hey VAXman-
>>[blah blah standard 'how come you not as rich as bill gates' claptrap removed]
>
>>So where is your software empire? I'm not out to make Microsoft
>>rich. They already are.
>
>Maybe VAXman has a shred of business ethics and morality...
You bet! I probably could have sold a number of my little ditties. I've
instead, chosen to publish them for the shared and mutual benefit of the
VMS community. And, as for business ethics and morality, I certainly do
have a whole lot more of them in my little pinky finger than a certain
someone (who I'll not name and it's not Mr. Gates) and his organization.
I've never been motivated by money! I grew up with not much and there-
fore, I don't lust for it other than that which is needed to pay the
bills -- of which I have my disproportionate share. I'm currently being
sued which has cost big-time for atty; my son, a year ago, had major
open-heart surgery; my wife has glaucoma and I'm slowly succumbing to
chronic renal failure. I work damn hard and take little odd-ball jobs
when I can to offset expenses (which has thrust me into a liesuit) but
I do not envy Mr. Gates nor his money. I'd rather envy his health --
he is in good health with all the $$$, isn't he?
I sit here, day after day, trying to make a living in the VMS market.
It ain't easy. But, I can leave this world with a little legacy and
the knowledge that, while not perfect, I've tried as best I can to do
the "Flushing Bay crawl" through this seething cesspool that has be-
come our industry.
Enough upon the soapbox.
>>I happen to admire Mr. Gates.
>
>I guess I could understand that if greed, underhandedness, double-crossing
>and anti-competitive trade practices are worthy of admiration.
The other night "Slick Willie" addressed the nation. "A computer in
every home" he said and I muted the volume to juxtapose, "and Micro-
Soft in every pocket". Imagine trying to find your car in a large
parking lot if the auto industry would have monopolized in the dir-
ections MicroSoft is going.
>>They all belong to Rush Limbaugh --so sue him.
>
>Q:What did bill gates say to the hotdog vendor
>A: Make me the one with everything.
ROTFL :) :) :)
VAXman- VAXman@TMESIS.COM
================================================================================
Archive-Date: Mon, 10 Feb 1997 19:07:00 +0100
Sender: owner-free-vms@lp.se
Message-ID: <c=US%a=_%p=msft%l=RED-70-MSG-970210180552Z-2334@INET-05-IMC.microsoft.com>
From: Chris Kaler <ckaler@microsoft.com>
Reply-To: Free-VMS@lp.se
To: "'Free-VMS@lp.se'" <Free-VMS@lp.se>
Subject: RE: A Potential Boost for Free-VMS
Date: Mon, 10 Feb 1997 10:05:52 -0800
Nice try. OpenVMS has NOT been released as shareware. It is still
a very viable business and Digital continues to make substantial
revenue on the product. I wouldn't expect sources to ever be released.
If you are referring to the A/D sources... well we are still trying to get
those released, but there will be restrictions around them and they
don't include things like clusters, security, or any of the kernel.
Christopher
-----Original Message-----
From: Gary.Gale@mail.sema.co.uk [SMTP:Gary.Gale@mail.sema.co.uk]
Sent: Monday, February 10, 1997 3:40 AM
To: Free-VMS@lp.se
Subject: A Potential Boost for Free-VMS
Hey all,
The discussions re to Mach or not to Mach and to Hurd and not to Hurd
seem to be getting a tad frantic - so I thought I'd interject with
something which _may_ give the whole project a boost (and may not - we'll
have to wait and see).
A colleague at the company I'm currently contracting for has just joined
us after a number of years with Digital here in the UK. The topic of
Free-VMS came up and I was informed that
"Digital has released about 60% of VMS as shareware"
Now from what I have managed to glean, this mythical 60% included both
kernel subsystems and user mode applications - in other words a sizeable
chunk of DCL and the innards of VMS itself.
I'm awaiting further information prior to passing it on to the group -
but and this is a big BUT (3 miles high, neon lights, you get the idea)
this could be just the boost needed to kick start some serious
development.
Watch this space ...
Gary Gale
SMS Unix & Gateways Team
Sema Group Telecoms, Reading, United Kingdom
+44 (0) 1734 64 1686 +44 (0) 973 188380
gary.gale@mail.sema.co.uk / gary.g@dial.pipex.com / 100742,400
================================================================================
Archive-Date: Mon, 10 Feb 1997 19:14:44 +0100
Sender: owner-free-vms@lp.se
Message-ID: <F14AAE60A252CF11B56100805F14DD0901D766D2@RED-70-MSG>
From: Chris Kaler <ckaler@microsoft.com>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: RE: Kernels, et al.
Date: Mon, 10 Feb 1997 10:13:07 -0800
So the last few days have seen a lot of discussion around the kernel.
Some really good ideas as well... so let me add some comments.
Hurd has some nice abstractions (although one might argue that they
came from Plan9), but isn't really suited (or wasn't the last time I
looked)
for multi-personality environments... still too much UNIX bias. Linux
also suffers this problem. And I would stick to kernels which are in
the
public domain.
So for a while I have been preaching that the best approach is to
do things in parallel and to gradually replace the system. What we
did was to use the BSD that came with Mach. We built our prototype
on top of UNIX to get it working and then slowly converted pieces to
work native (removed the UNIX dependencies). We never finished,
but we were able to show a working system very quickly, and we were
able to leverage a lot of the UNIX code in order to have UNIX and VMS
apps running together. I would strongly recommend this approach.
The HURD idea for generic file system handles is a good one. It is
similar to the Win32 concept of monikers. If you aren't familiar with
either of them, you should read up. I would advocate taking these
ideas even further and have generic objects so that everything is
plug-and-play as sub-classes of a common type.
Christopher
================================================================================
Archive-Date: Mon, 10 Feb 1997 20:20:11 +0100
Sender: owner-free-vms@lp.se
Subject: Re: Kernels and VMS
To: Free-VMS@lp.se
Date: Mon, 10 Feb 1997 14:20:05 -0500 (EST)
From: "christopher f. chiesa" <lvt-cfc@servtech.com>
Reply-To: Free-VMS@lp.se
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Message-ID: <32ff74e50.2aa2@cyber2.servtech.com>
Content-Transfer-Encoding: 7bit
Hey, I've GOT it! This wraps up the whole Linux/Mach/whatever platform
debate once and for all, AND finishes the Free-VMS project way ahead of
schedule, all in one fell swoop.
Here's what ya do:
Find someone who's got some sort of VAX/VMS (or, eventually,
OpenVMS Alpha) system they're "just going to THROW OUT," and
get them to give it to you (this is the "free" part). Take
it home, and use it (this is the "VMS" part).
Problem solved: you can run VMS at home and it's free. What am I missing?
(Darn, my system clock must be screwed up; I could've SWORN it was
1-APR...)
Chris Chiesa
================================================================================
Archive-Date: Mon, 10 Feb 1997 21:20:33 +0100
Sender: owner-free-vms@lp.se
Message-ID: <01BC1790.00AA46E0@PRO4>
From: "John F. Roden" <j.roden@procyon.ie>
Reply-To: Free-VMS@lp.se
To: "'Free-VMS@lp.se'" <Free-VMS@lp.se>
Subject: RE: Kernels, et al.
Date: Mon, 10 Feb 1997 20:20:31 -0000
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="---- =_NextPart_000_01BC1790.00AD5420"
------ =_NextPart_000_01BC1790.00AD5420
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Chris Kaler wrote:-
> The HURD idea for generic file system handles is a good one. It is
> similar to the Win32 concept of monikers. If you aren't familiar with
> either of them, you should read up. I would advocate taking these
> ideas even further and have generic objects so that everything is
> plug-and-play as sub-classes of a common type.
This is what I was weakly trying to suggest! I think we should have 3 layers
something like this:
1. Mach
2. Objects-support
3. VMS
The objects layer would provide a set of services to allow every type of
program (drivers, queue handlers, directory handers, files ...) to be treated
as objects in a common name space. Each object would present one
or more API's which would allow other objects to make use of it. The linking
of objects is the major challenge -- if it is really general it would allow everything
from memory sharing to RPC between objects. By defining a common object model
we could probably simplify a lot of the system services development.
The VMS layer would translate the facilities of the objects layer into a
set of API's which would mimic the existing VMS system services. However,
the underlying objects layer would let us extend the services in the future.
I'll try to write an outline design for the object approach.
John Roden
------ =_NextPart_000_01BC1790.00AD5420
Content-Type: application/ms-tnef
Content-Transfer-Encoding: base64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------ =_NextPart_000_01BC1790.00AD5420--
================================================================================
Archive-Date: Tue, 11 Feb 1997 01:48:48 +0100
Sender: owner-free-vms@lp.se
Date: Mon, 10 Feb 1997 16:48:42 -0800
From: Tony Konashenok <tonyk@sseos.lbl.gov>
Reply-To: Free-VMS@lp.se
Message-ID: <199702110048.QAA03611@sseos.lbl.gov>
To: free-vms@lp.se
Subject: RE: Kernels, et al.
Chris Kaler wrote:
$! So for a while I have been preaching that the best approach is to
$! do things in parallel and to gradually replace the system. What we
$! did was to use the BSD that came with Mach. We built our prototype
$! on top of UNIX to get it working and then slowly converted pieces to
$! work native (removed the UNIX dependencies). We never finished,
$! but we were able to show a working system very quickly, and we were
$! able to leverage a lot of the UNIX code in order to have UNIX and VMS
$! apps running together. I would strongly recommend this approach.
<Asbestos suit on>
This has a strong foundation, except for one thing: goals are to be
achieved earlier at the cost of significant performance penalty, and
maybe some constraints which wouldn't be otherwise imposed. Conversion
to the high-performance version is to happen at a later date. Having
worked in three understaffed, underfunded projects held together by
sheer willpower of the employees, I happen to know that the time frame
for a performance release is "hardly ever". Even though we may not even
have any other choice than do as Chris suggested, I would still urge
everyone to be vewy, vewy caweful about it. Consider historical
precedent: the very reason that VMS is so stable, reliable, secure, etc.
is that it was a commercial project from the very beginning, and the
corporate culture of DEC was one of good programming. Unix, on the
other hand, even though started by good programmers, was not originally
intended to be a commercial system, and so suffered from sloppy attitude,
especially in the area of the user interface. Another example... one
unnamed company in the state of Washington is so dedicated to marketing
that it made sloppy coding its corporate culture (no offense meant to
Chris). Actually, I perceive the latter two cases as one of the major
driving forces behind the Free-VMS project. So, do we want to write good,
efficient, stable code, or do we want to write good, efficient stable
code? One possible approach could be to start writing TWO releases at
once, sharing the top-level API code and all utilities and applications,
but one aimed at achieving most features in the shortest time, and the
other aimed at achieving most performance and reliability, even if it
costs some features. (Actually, I am already doing some study on the
second variant, but the results are too preliminary to report).
Tony Konashenok
================================================================================
Archive-Date: Tue, 11 Feb 1997 17:49:42 +0100
Sender: owner-free-vms@lp.se
Message-ID: <c=US%a=_%p=msft%l=RED-70-MSG-970211164716Z-6203@INET-04-IMC.microsoft.com>
From: Chris Kaler <ckaler@microsoft.com>
Reply-To: Free-VMS@lp.se
To: "'Free-VMS@lp.se'" <Free-VMS@lp.se>
Subject: RE: Kernels, et al.
Date: Tue, 11 Feb 1997 08:47:16 -0800
I'll see your flame-thrower and raise you one ;-)
Your point is valid, but I believe the correlation to my statement is weak.
My first reaction is to say that our approach yielded performance results
that often beat the OpenVMS kernel. The approach itself does not
introduce performance penalties. What is does is provide a development,
runtime, and debugging environment that is stable. Have you ever tried
to write something like this from scratch. When you try to debug you
have 47 variables and no idea where to start. It is important to have
some basic foundation to build on. With UNIX on Mach you get the
C RTL, boot code, paging, and a file system. I would argue that in writing
your own versions of those you will not see great performance increases
because the people working on Mach have tried to make it go fast. And
the 5 billion people hacking on UNIX around the world have also tried
to make it go fast. And guess what, they did a pretty good job.
You point about writing good code is right on the money, but I don't
agree that it must be a "commercial" venture. If you focus on building
a high-quality product with strong performance, that is what you will
build. Being a commercial product requires you to meet those goals
or die, whereas a private venture can survive even if the performance
isn't fabulous. However, if you try to make this a commercial venture
then you will probably see brick walls appearing before your very
eyes... and DEC can spend more on lawyers than you can.
So let's build something that has strong performance, is free for anyone
to use or modify, and let's do it in a sane and rationale manner so that
the people writing code don't pull all their hair out.
Chris
-----Original Message-----
From: Tony Konashenok [SMTP:tonyk@sseos.lbl.gov]
Sent: Monday, February 10, 1997 4:49 PM
To: free-vms@lp.se
Subject: RE: Kernels, et al.
Chris Kaler wrote:
$! So for a while I have been preaching that the best approach is to
$! do things in parallel and to gradually replace the system. What we
$! did was to use the BSD that came with Mach. We built our prototype
$! on top of UNIX to get it working and then slowly converted pieces to
$! work native (removed the UNIX dependencies). We never finished,
$! but we were able to show a working system very quickly, and we were
$! able to leverage a lot of the UNIX code in order to have UNIX and VMS
$! apps running together. I would strongly recommend this approach.
<Asbestos suit on>
This has a strong foundation, except for one thing: goals are to be
achieved earlier at the cost of significant performance penalty, and
maybe some constraints which wouldn't be otherwise imposed. Conversion
to the high-performance version is to happen at a later date. Having
worked in three understaffed, underfunded projects held together by
sheer willpower of the employees, I happen to know that the time frame
for a performance release is "hardly ever". Even though we may not even
have any other choice than do as Chris suggested, I would still urge
everyone to be vewy, vewy caweful about it. Consider historical
precedent: the very reason that VMS is so stable, reliable, secure, etc.
is that it was a commercial project from the very beginning, and the
corporate culture of DEC was one of good programming. Unix, on the
other hand, even though started by good programmers, was not originally
intended to be a commercial system, and so suffered from sloppy attitude,
especially in the area of the user interface. Another example... one
unnamed company in the state of Washington is so dedicated to marketing
that it made sloppy coding its corporate culture (no offense meant to
Chris). Actually, I perceive the latter two cases as one of the major
driving forces behind the Free-VMS project. So, do we want to write good,
efficient, stable code, or do we want to write good, efficient stable
code? One possible approach could be to start writing TWO releases at
once, sharing the top-level API code and all utilities and applications,
but one aimed at achieving most features in the shortest time, and the
other aimed at achieving most performance and reliability, even if it
costs some features. (Actually, I am already doing some study on the
second variant, but the results are too preliminary to report).
Tony Konashenok
================================================================================
Archive-Date: Tue, 11 Feb 1997 20:17:57 +0100
Sender: owner-free-vms@lp.se
Date: Tue, 11 Feb 1997 11:17:31 -0800
From: Tony Konashenok <tonyk@sseos.lbl.gov>
Reply-To: Free-VMS@lp.se
Message-ID: <199702111917.LAA19869@sseos.lbl.gov>
To: free-vms@lp.se
Subject: RE: Kernels, et al.
Chris Kaler:
$! However, if you try to make this a commercial venture
$! then you will probably see brick walls appearing before your very
$! eyes... and DEC can spend more on lawyers than you can.
No, we are doing a free project. What I meant was that the attitude of the
participants should be the same as in a commercial project -- meet the
goals or die, just as you said. Better not die but meet the goals...
$! So let's build something that has strong performance, is free for anyone
$! to use or modify, and let's do it in a sane and rationale manner so that
$! the people writing code don't pull all their hair out.
I'll second that.
Tony Konashenok
================================================================================
Archive-Date: Wed, 12 Feb 1997 02:15:25 +0100
Sender: owner-free-vms@lp.se
Date: Tue, 11 Feb 1997 20:14:20 EST
From: "Brian Schenkenberger, VAXman-" <system@TMESIS.COM>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Message-ID: <009AFBE0.A0DBFEEE.3@TMESIS.COM>
Subject: RE: Kernels, et al.
Tony Konashenok <tonyk@sseos.lbl.gov> sez:
>Chris Kaler:
>
>$! However, if you try to make this a commercial venture
>$! then you will probably see brick walls appearing before your very
>$! eyes... and DEC can spend more on lawyers than you can.
I got news for ya. DEC may have more lawyers than you can muster but
they don't do anything with 'em e'cept keep 'em well fed on the Shark
bait of all legal sharks -- $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
DEC already knows of people reselling their product but does nothing!
Don't it make you mad? We've got to pay big bucks for VMS if we want
to run VMS -- hence, is this not the thrust behind a FreeVMS project?
If DEC went after the jokers stealing their code, they could possibly
come to face with creating a "hobbyist" class VMS license and perhaps,
get some more enthusiasm into the VMS marketplace to their benefit.
O! But nay! They're contented to keep baiting away the sharks and
engorging their stomachs with paper notes donned 'e pluribus unim'!
BTW, I'd be careful in this project simply because of the shark in-
fested waters. There's a lot of software out there that's protected
by patent -- VMS I'm sure has some components. I spoke with an eng'r
recently that hinted that DEC had submitted for patent, some of the
new 64 bit mmg mechanism; I'm sure there's probably more in VMS.
VAXman- VAXman@TMESIS.COM
================================================================================
Archive-Date: Thu, 13 Feb 1997 14:36:16 +0100
Sender: owner-free-vms@lp.se
MIME-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Date: Thu, 13 Feb 1997 14:36:12 MET
Message-ID: <13059.9948.271759.feedmail.pl5.LEVITTE@devil.bofh.se>
From: Richard Levitte - VMS Whacker <levitte@lp.se>
Reply-To: Free-VMS@lp.se
X-MX-Warning: Warning -- Invalid "To" header.
To: free-vms
remove stefan@BBFVLC.TU-VARNA.ACAD.BG
================================================================================
Archive-Date: Thu, 13 Feb 1997 14:39:54 +0100
Sender: owner-free-vms@lp.se
MIME-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Date: Thu, 13 Feb 1997 14:39:48 MET
Message-ID: <13059.10164.271759.feedmail.pl5.LEVITTE@devil.bofh.se>
From: