the document describing the system service transfer vector

[hb]

Hi Camiel,

https://vmsbug.eng.vmssoftware.com/mediawiki/index.php/File:ModularExecutive.docx
is the document. As I said, I think it should be updated.

As explained in the document, the linker leaves 64 bytes for writing the
actual transfer vector in the system service loader. The additional code
to touch the stack should fit in.

Hartmut

[hb]

Sorry, please ignore, sent to wrong recipient.

[Andy Burns]

hb wrote:

> https://vmsbug.eng.vmssoftware.com/mediawiki/index.php/File:ModularExecutive.docx

not WPS-Plus?

[Stephen Hoffman]

Nope. VSI upgraded to DECwriteX a while back.


--
Pure Personal Opinion | HoffmanLabs LLC

[VAXman-]

That URL is invalid.

--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG

I speak to machines with the voice of humanity.

[Andy Burns]

VAXman wrote:

> Andy Burns writes:
>
>> hb wrote:
>>
>>> https://vmsbug.eng.vmssoftware.com/mediawiki/index.php/File:ModularExecutive.docx
>>
>> not WPS-Plus?
>
> That URL is invalid.

Maybe it works internally?

[VAXman-]

Even if it DID work, what would I do with it? No WEENDOZE to read it. What's
wrong with PDF? Where the P is Portable.

[Single Stage to Orbit]

On Mon, 2022-09-19 at 18:57 +0000, VAX...@SendSpamHere.ORG wrote:
> > > https://vmsbug.eng.vmssoftware.com/mediawiki/index.php/File:ModularExecutive.docx
> > >  
> >
> > not WPS-Plus?
>
> That URL is invalid.

It's valid. You can't access it from outside VSI.
--
Tactical Nuclear Kittens

[Bill Gunshannon]

On 9/19/22 17:40, VAX...@SendSpamHere.ORG wrote:
> In article <jorsn3...@mid.individual.net>, Andy Burns <use...@andyburns.uk> writes:
>> VAXman wrote:
>>
>>> Andy Burns writes:
>>>
>>>> hb wrote:
>>>>
>>>>> https://vmsbug.eng.vmssoftware.com/mediawiki/index.php/File:ModularExecutive.docx
>>>>
>>>> not WPS-Plus?
>>>
>>> That URL is invalid.
>>
>> Maybe it works internally?
>
> Even if it DID work, what would I do with it? No WEENDOZE to read it. What's
> wrong with PDF? Where the P is Portable.
>

Don't need Windows to read a docx file. It really has nothing to do
with Windows. I read them on Linux and MacOS all the time.

bill

[Arne Vajhøj]

It just requires something that reads OOXML.

MS Office (Windows) and the various LibreOffice/OpenOffice variants
(Windows, Linux, macOS, *BSD) can read and write.

Lotus Notes (Windows, Linux, macOS, i), WordPerfect (Windows) and
Google Docs (web) can read.

But I don't think anything on VMS can read them.

Arne

[Bill Gunshannon]

Who's fault is that?

bill

[David Jones]

On Monday, September 19, 2022 at 7:56:35 PM UTC-4, Arne Vajhøj wrote:
> On 9/19/2022 6:55 PM, Bill Gunshannon wrote:
> > Don't need Windows to read a docx file. It really has nothing to do
> > with Windows. I read them on Linux and MacOS all the time.
> It just requires something that reads OOXML.
>
> MS Office (Windows) and the various LibreOffice/OpenOffice variants
> (Windows, Linux, macOS, *BSD) can read and write.
>

> But I don't think anything on VMS can read them.
>

I read .xlsx files on VMS, mapping cell ranges on sheets to SQLite virtual
tables. It's best if you don't get too fancy with text attributes. The ECMA
specification is really abstruse.

[Arne Vajhøj]

Is it anybodys fault?

Word processors is a desktop application.

That has not been VMS'es domain for decades.

Arne

[Arne Vajhøj]

Good point.

I was really thinking about word processors.

But of course the XML can be read and parsed.

If one like stuff that starts with J then POI
can do it.

An SQLite virtual table sounds pretty cool
though!

Is it available?

Arne

[David Jones]

On Monday, September 19, 2022 at 9:24:32 PM UTC-4, Arne Vajhøj wrote:
> An SQLite virtual table sounds pretty cool
> though!
>
> Is it available?
>

It's on sourceforge: https://sourceforge.net/projects/vms-ports/files/SQLITE3/
Look for excel_convert_004.zip. The extension uses the libz and libbz2 libraries,
not included in the kit, to decompress the data in .xlsx files.

[Bill Gunshannon]

And who's fault is that? :-)


bill

[Arne Vajhøj]

Who made the decision for VMS to go after the
expensive server market and not the cheap
desktop market?

I suspect Ken Olsen is ultimate to blame or get
credit for the decision.

But whether one think that was the wrong or the right
decision, then that doesn't change where we are today.

Arne

[Bill Gunshannon]

In Ken Olsen's day VMS only ran on proprietary DEC hardware which
meant DEC would have had to make the desktop hardware as well as
the software. That's not true today. Maybe it's time to re-look
at the situation with an eye towards a much more secure desktop
option. especially when one considers that the majority of attacks
start at the desktop. :-)

bill

[Phillip Helbig (undress to reply)]

In article <jou0cl...@mid.individual.net>, Bill Gunshannon

<bill.gu...@gmail.com> writes:

> In Ken Olsen's day VMS only ran on proprietary DEC hardware which
> meant DEC would have had to make the desktop hardware as well as
> the software. That's not true today. Maybe it's time to re-look
> at the situation with an eye towards a much more secure desktop
> option. especially when one considers that the majority of attacks
> start at the desktop. :-)

Right. I just have to convince Elon to pay for a modern web browser for
VMS.

[John Dallman]

In article <tgcoj9$1fst$1...@gioia.aioe.org>,
hel...@asclothestro.multivax.de (Phillip Helbig (undress to reply))
wrote:

> Right. I just have to convince Elon to pay for a modern web
> browser for VMS.

Or set to work on the open-source Chromium browser, probably starting
with third-party libraries that it uses, most of which would be
independently useful. It's a lot of work, but there's nothing impossible
about it.

John

[Simon Clubley]

Interesting.

I didn't know there were Jovial libraries available for such things. :-)

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.

[Bill Gunshannon]

Unless things have changed modern web browsers are available
Open Source. All it would take was someone who knew VMS
development to do the port. After, of course, someone did
a port of a modern version of X11 (also opensource). :-)

bill

[Arne Vajhøj]

It would probably be hard to find a 100% closed source web
browser today.

:-)

But they are pretty big.

It is more than the browser and X.

It would need GTK+ or Qt on top of X.

It would need a JavaScript engine.

It would need uptodate versions of a bunch of graphic
and video libraries.

25 years ago Mosaic and Lynx on VMS was sort of uptodate.

But since then the size of web browsers has increased
by a factor 10 or more and the willingness in VMS
community to work on open source has decreased by
a factor 10 or more.

I don't think it is realistic.

Arne

[Simon Clubley]

On 2022-09-20, Arne Vajhøj <ar...@vajhoej.dk> wrote:
> On 9/20/2022 2:17 PM, Bill Gunshannon wrote:
>>
>> Unless things have changed modern web browsers are available
>> Open Source.  All it would take was someone who knew VMS
>> development to do the port.  After, of course, someone did
>> a port of a modern version of X11 (also opensource).  :-)
>
> It would probably be hard to find a 100% closed source web
> browser today.
>
>:-)
>
> But they are pretty big.
>
> It is more than the browser and X.
>
> It would need GTK+ or Qt on top of X.
>
> It would need a JavaScript engine.
>
> It would need uptodate versions of a bunch of graphic
> and video libraries.
>

Also add:

Modern version of C++.

Depending on browser, maybe a Rust port as well.

A modern build system.

Emulation of any missing or incomplete Unix APIs that the browser needs.

It is also unclear to me if the version of X running under VMS would
be modern enough to support such a port.

[Bill Gunshannon]

Most people in the industry today don't think the existence of VMS is
realistic, but here we are.

I don't expect to see it happen, but I would really be happy if it did.

bill

[Bill Gunshannon]

On 9/20/22 14:40, Simon Clubley wrote:
> On 2022-09-20, Arne Vajhøj <ar...@vajhoej.dk> wrote:
>> On 9/20/2022 2:17 PM, Bill Gunshannon wrote:
>>>
>>> Unless things have changed modern web browsers are available
>>> Open Source.  All it would take was someone who knew VMS
>>> development to do the port.  After, of course, someone did
>>> a port of a modern version of X11 (also opensource).  :-)
>>
>> It would probably be hard to find a 100% closed source web
>> browser today.
>>
>> :-)
>>
>> But they are pretty big.
>>
>> It is more than the browser and X.
>>
>> It would need GTK+ or Qt on top of X.
>>
>> It would need a JavaScript engine.
>>
>> It would need uptodate versions of a bunch of graphic
>> and video libraries.
>>
>
> Also add:
>
> Modern version of C++.
>
> Depending on browser, maybe a Rust port as well.
>
> A modern build system.
>
> Emulation of any missing or incomplete Unix APIs that the browser needs.
>
> It is also unclear to me if the version of X running under VMS would
> be modern enough to support such a port.

Thus the reason I said, "After, of course, someone did

a port of a modern version of X11 (also opensource)."

bill