Has anyone collected the Multics source listings into a zip or tar type
of archive? I find viewing
http://web.mit.edu/multics-history/source/Multics_Internet_Server/Multics_sources.html
in the browser annoying and hard to search.

Regards,
Kevin

In article <sd5npb$pvs$1@842ffb22-07e1-11e5-a459-00266cf00584.csiph.com>,
Kevin Bowling <kevin.bowling@kev009.com> wrote:
>Has anyone collected the Multics source listings into a zip or tar type
>of archive? I find viewing
>http://web.mit.edu/multics-history/source/Multics_Internet_Server/Multics_sources.html
>in the browser annoying and hard to search.
>
>Kevin

Kevin,

No one else has responded, so let me see if I can help.

Please excuse me for "man-splainin'" herein, but I don't know your
background ... Also, I'm going to simplify things here by asserting
some things that I am only assuming.

First, you could, of course, relatively easily make such a copy using,
for example, wget.

Second, I have such an archive copy, but it's on my own
(non-net-accessible) machine and not in zip/tar form (although that
would be trivial to create).

But first, some questions:

Are you sure that the MIT archive is what you want? It's incomplete,
and the structure is painful. (It's Multics format adapted to UNIX
format and then forced into HTTP limitations.) I had to do a lot of
manipulation on my copy to make it useful to me, so my copy is not a
true copy.

In particular, are you aware that a group of dedicated hackers has
taken the original Multics system tapes (from which the MIT archive is
derived) and recreated a working system (almost) fully derived from
(updated) source code. (There is source code for the entire base
system, but some optional components, such as the C compiler, are
binary only.) Are you aware that you too could run Multics on your
local system? Are you aware that there is a Multics running on the
net that provides guest access?

(I'm too lazy at the moment to look up where the current web site for
this is, but I'm almost positive that it's referenced on the front
page of www.multicians.org.)

I'm not aware of a web-accessible copy of that later source (it's in
its "proper" place within a running Multics system). I do, however,
have a pieced-together copy of that source on my local system. It's
slightly different in structure (e.g., all Multics archive files have
been replaced by a directory containing individual component files).
Also, it's the source for Multics Release 12.6f, while 12.7 is in the
final stages of being released (or might have actually been released
already). (I'm actually running a beta version of 12.7 locally.)

I have limited options for storing files accessible on the net. If it
would be useful to you I could investigate my storage and data
transfer limits, and possibly upload one or the other of my archives.
If somehow it's the MIT archive that is really of interest, please let
me know if the real "listings" (the files with a ".list" suffix) are
really of use to you. The MIT archive is about 1GB and the listings
make up about half of that -- and I find them almost useless for my
purposes.

Also, is DropBox an option for you? (I'm a member, but I'd have to
learn how to upload to it.)

- dmw

On 7/21/21 11:19 AM, Douglas Wells wrote:
> In article <sd5npb$pvs$1@842ffb22-07e1-11e5-a459-00266cf00584.csiph.com>,
> Kevin Bowling <kevin.bowling@kev009.com> wrote:
>> Has anyone collected the Multics source listings into a zip or tar type
>> of archive? I find viewing
>> http://web.mit.edu/multics-history/source/Multics_Internet_Server/Multics_sources.html
>> in the browser annoying and hard to search.
>>
>> Kevin
>
> Kevin,
>
> No one else has responded, so let me see if I can help.
>
> Please excuse me for "man-splainin'" herein, but I don't know your
> background ... Also, I'm going to simplify things here by asserting
> some things that I am only assuming.
>
> First, you could, of course, relatively easily make such a copy using,
> for example, wget.
>
> Second, I have such an archive copy, but it's on my own
> (non-net-accessible) machine and not in zip/tar form (although that
> would be trivial to create).
>
> But first, some questions:
>
> Are you sure that the MIT archive is what you want? It's incomplete,
> and the structure is painful. (It's Multics format adapted to UNIX
> format and then forced into HTTP limitations.) I had to do a lot of
> manipulation on my copy to make it useful to me, so my copy is not a
> true copy.
>
> In particular, are you aware that a group of dedicated hackers has
> taken the original Multics system tapes (from which the MIT archive is
> derived) and recreated a working system (almost) fully derived from
> (updated) source code. (There is source code for the entire base
> system, but some optional components, such as the C compiler, are
> binary only.) Are you aware that you too could run Multics on your
> local system? Are you aware that there is a Multics running on the
> net that provides guest access?
>
> (I'm too lazy at the moment to look up where the current web site for
> this is, but I'm almost positive that it's referenced on the front
> page of www.multicians.org.)
>

Yeah I have seen I think all of the above in my several hours of research.

I'll explain what I want to do: run the tool ripgrep (a modern lighting
fast grep style program) to explore the structure and concepts and find
parts I want to read, and open it up in my text programs of choice for
longer form readings. I don't really intend to modify it or compile it
so it can be incomplete or an imperfect projection to different FS
semantics as you are leading!

> I'm not aware of a web-accessible copy of that later source (it's in
> its "proper" place within a running Multics system). I do, however,
> have a pieced-together copy of that source on my local system. It's
> slightly different in structure (e.g., all Multics archive files have
> been replaced by a directory containing individual component files).
> Also, it's the source for Multics Release 12.6f, while 12.7 is in the
> final stages of being released (or might have actually been released
> already). (I'm actually running a beta version of 12.7 locally.)
>
> I have limited options for storing files accessible on the net. If it
> would be useful to you I could investigate my storage and data
> transfer limits, and possibly upload one or the other of my archives.
> If somehow it's the MIT archive that is really of interest, please let
> me know if the real "listings" (the files with a ".list" suffix) are
> really of use to you. The MIT archive is about 1GB and the listings
> make up about half of that -- and I find them almost useless for my
> purposes.
>
> Also, is DropBox an option for you? (I'm a member, but I'd have to
> learn how to upload to it.)

Sure, any means you have available to transfer are appreciated. I
believe my email address can take up to a 25M attachment too.

> - dmw
>

Kevin Bowling <kevin.bowling@kev009.com> writes:
> ... run the tool ripgrep (a modern lighting
> fast grep style program)

I've been using "ack". Thanks for the pointer, "ripgrep" does appear to be
significantly faster!

Andy Valencia
Home page: https://www.vsta.org/andy/
To contact me: https://www.vsta.org/contact/andy.html

Kevin,

In article <sdivq0$15hu$1@842ffb22-07e1-11e5-a459-00266cf00584.csiph.com>,
Kevin Bowling <kevin.bowling@kev009.com> wrote:
>On 7/21/21 11:19 AM, Douglas Wells wrote:
>> In article <sd5npb$pvs$1@842ffb22-07e1-11e5-a459-00266cf00584.csiph.com>,
>> Kevin Bowling <kevin.bowling@kev009.com> wrote:
>>> Has anyone collected the Multics source listings into a zip or tar type
>>> of archive? I find viewing
>>> http://web.mit.edu/multics-history/source/Multics_Internet_Server/Multics_sources.html
>>> in the browser annoying and hard to search.

>> [...] I have such an archive copy, but it's on my own
>> (non-net-accessible) machine and not in zip/tar form (although that
>> would be trivial to create).

>I'll explain what I want to do: run the tool ripgrep (a modern lighting
>fast grep style program) to explore the structure and concepts and find
>parts I want to read, and open it up in my text programs of choice for
>longer form readings. I don't really intend to modify it or compile it
>so it can be incomplete or an imperfect projection to different FS
>semantics as you are leading!

OK, so I've created an online archive of MR12.6f (released 2017). It's in:
<URL:ftp://ftp.drivehq.com/dmwells22/mxprogs/multics-source/README.html>

I suggest you start with the README file at either:
<URL:ftp://ftp.drivehq.com/dmwells22/mxprogs/multics-source/README.md>
or
<URL:ftp://ftp.drivehq.com/dmwells22/mxprogs/multics-source/README.html>

Please let me know of any problems as I've done only minimal testing on it.

-dmw

In article <1627356187.78503@nowhere.invalid>,
Douglas Wells <nr201504@gmail.com> wrote:
>In article <sdivq0$15hu$1@842ffb22-07e1-11e5-a459-00266cf00584.csiph.com>,
>Kevin Bowling <kevin.bowling@kev009.com> wrote:
>>[...]
>>I'll explain what I want to do: run the tool ripgrep (a modern lighting
>>fast grep style program) to explore the structure and concepts and find
>>parts I want to read, and open it up in my text programs of choice for
>>longer form readings. I don't really intend to modify it or compile it
>>so it can be incomplete or an imperfect projection to different FS
>>semantics as you are leading!
>
>OK, so I've created an online archive of MR12.6f (released 2017). It's in:
><URL:ftp://ftp.drivehq.com/dmwells22/mxprogs/multics-source/README.html>
>
>I suggest you start with the README file at either:
><URL:ftp://ftp.drivehq.com/dmwells22/mxprogs/multics-source/README.md>
>or
><URL:ftp://ftp.drivehq.com/dmwells22/mxprogs/multics-source/README.html>
>
>Please let me know of any problems as I've done only minimal testing on it.

Very cool; thank you for doing that. For convenience and discoverability,
I've extracted this and pushed it to github:

https://github.com/dancrossnyc/multics

I'm kind of surprised it's not already there, but I didn't find it in
a cursory examination.

- Dan C.

On 7/27/21 4:16 AM, Dan Cross wrote:
> In article <1627356187.78503@nowhere.invalid>,
> Douglas Wells <nr201504@gmail.com> wrote:
>> In article <sdivq0$15hu$1@842ffb22-07e1-11e5-a459-00266cf00584.csiph.com>,
>> Kevin Bowling <kevin.bowling@kev009.com> wrote:
>>> [...]
>>> I'll explain what I want to do: run the tool ripgrep (a modern lighting
>>> fast grep style program) to explore the structure and concepts and find
>>> parts I want to read, and open it up in my text programs of choice for
>>> longer form readings. I don't really intend to modify it or compile it
>>> so it can be incomplete or an imperfect projection to different FS
>>> semantics as you are leading!
>>
>> OK, so I've created an online archive of MR12.6f (released 2017). It's in:
>> <URL:ftp://ftp.drivehq.com/dmwells22/mxprogs/multics-source/README.html>
>>
>> I suggest you start with the README file at either:
>> <URL:ftp://ftp.drivehq.com/dmwells22/mxprogs/multics-source/README.md>
>> or
>> <URL:ftp://ftp.drivehq.com/dmwells22/mxprogs/multics-source/README.html>
>>
>> Please let me know of any problems as I've done only minimal testing on it.
>
> Very cool; thank you for doing that. For convenience and discoverability,
> I've extracted this and pushed it to github:
>
> https://github.com/dancrossnyc/multics
>
> I'm kind of surprised it's not already there, but I didn't find it in
> a cursory examination.
>
> - Dan C.
>

Thanks Doug and Dan! This will result in easier discoverability and
more young engineers being able to study the code.

Regards,
Kevin

You can download the entire MR12.3 source tree from here:

https://drive.google.com/file/d/1DUMripAHXAyDjAdqH05v011gw3HgiJ4z/view?usp=sharing

It contains useless binaries as well; but .archive files are readable in an editor. Olin's mxload
should be able to split them up, but it's tricky to use.

-Dai