https://softwarepreservation.computerhistory.org/modula3/

[pmcjones]

https://softwarepreservation.computerhistory.org/modula3/ is a web site dedicated to recording the history of the Modula-3 programming language, its implementations, and its users. I got the idea for this several years ago, but didn't make much progress until last fall. It's hosted by the Computer History Museum, as part of the Software Preservation Group, which began in 2003, triggered by a 2002 email blast by Grady Booch.

Let me know if you have suggestions for the web site. I was hoping to find source code for GNU Modula-3, but that seems unlikely now. Also, I'm interested in work that may have taken place at Xerox PARC involving M3 and their Portable Common Runtime. Also, there's the work that Mary Fernandez did involving link-time optimization of opaque types.

[demoitem]

That is very interesting. I wonder what happened to the original C backend code? Lost I imagine. It must have been called
directly from the front-end since I gather they didnt have the middle layer back then. I would also like to read that
link-time optimization paper if you ever find it.

[pmcjones]

As the web site say, "The original version generated C and invoked GCC to compile it, but starting with version 3.0 it was modified to produce an intermediate language that could drive different backends.” There is source code online for versions 1.5, 2.09, 2.10, and 2.11. At some point the releases started including enough generated code to allow bootstrapping onto another platform. For example, 2.10 includes the m3.2.10/compiler/boot-DS3100/ directory with generated C code. And 2.11 has a whole family of bootstrap directories — see https://softwarepreservation.computerhistory.org/modula3/archive/DEC_SRC/release-2.11-unpacked/ .

Mary Fernandez's work on link-time optimization of opaque types includes:

Fernandez, Mary F.
“Simple and Effective Link-Time Optimization of Modula-3 Programs.” PLDI ’95, ACM SIGPLAN Conference on Programming Language Design and Implementation, 1995.

Fernandez, Mary F.
“Simple and Effective Link-Time Optimization of Modula-3 Programs.” Princeton University Technical Report TR‑474‑94, November 1994.

Fernandez, Mary.
“Link-Time Optimization of Modula-3.” Article in Threads (the Modula‑3 newsletter), issue 2, mid‑1990s.

Fernandez, Mary F.
“The Implementation of mlcc and mld: A Machine-Independent Compiler and a Retargetable, Optimizing Linker.” Princeton University Technical Report TR‑490‑95, 1995.

[pmcjones]

I've updated the web site to discuss Mary Fernández's work and link to her papers:

Princeton M3

[demoitem]

Thanks for that. I had a look at one of her papers and she seems to be way ahead of her time. I've been doing some work on the llvm backend and they have LTO but it was only implemented about 4 -5 years ago I think. If you ever have to compile llvm you are in for a rude shock as far as build times go compared to modula3. It's a nightmare. Anyway congrats on the website. I'm still checking out the references.

…

On Tue, 17 Feb 2026 at 13:48, Paul McJones ***@***.***> wrote: I've updated the web site to discuss Mary Fernández's work and link to her papers: Princeton M3 <https://softwarepreservation.computerhistory.org/modula3/#Princeton> — Reply to this email directly, view it on GitHub <#1221 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABALZX44SMZSFI32MOM2V3T4MJ6QFAVCNFSM6AAAAACUU6MTDSVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTKOBSHE4DKMY> . You are receiving this because you commented.Message ID: ***@***.***>

[olaf-wagner]

On 10 Feb 2026, at 20:16, Paul McJones wrote: https://softwarepreservation.computerhistory.org/modula3/ is a web site dedicated to recording the history of the Modula-3 programming language, its implementations, and its users. I got the idea for this several years ago, but didn't make much progress until last fall. It's hosted by the Computer History Museum, as part of the Software Preservation Group, which began in 2003, triggered by a 2002 email blast by Grady Booch. Let me know if you have suggestions for the web site. I was hoping to find source code for GNU Modula-3, but that seems unlikely now. Also, I'm interested in work that may have taken place at Xerox PARC involving M3 and their Portable Common Runtime. Also, there's the work that Mary Fernandez did involving link-time optimization of opaque types.

Hi again Paul McJones, thanks for all the work you have invested in the history of Modula-3. I seem to remember that I still owe you some information, but have forgotten what exactly I wanted to dig out. I'm also not sure if Mike ***@***.***) has sent you the source code archives we still have somewhere. However, when tidying up some old shelves here in our office, I found an original CM3 Reactor CD from Critical Mass I got from Farshad Nayeri long ago. I've attached a photo just in case you are curious. I haven't tried to read the CD yet, because my computer doesn't have a suitable drive for that. Perhaps Mike can try it. Regards from Berlin, Olaf -- Olaf Wagner -- elego Software Solutions GmbH -- http://www.elegosoft.com Gustav-Meyer-Allee 25 / Gebäude 12, 13355 Berlin, Germany phone: +49 30 23 45 86 96 mobile: +49 163 52 11 579 fax: +49 30 23 45 86 95 Geschäftsführer: Olaf Wagner | Sitz: Berlin Handelregister: Amtsgericht Charlottenburg HRB 77719 | USt-IdNr: DE163214194

[VictorMiasnikov]

Attempt N2 2026-02-19 |
Attempt N2 2026-02-19 |> Regards from Berlin, Olaf -- Olaf Wagner -- elego Software Solutions GmbH
Attempt N2 2026-02-19 |
Attempt N2 2026-02-19 |I try solve ( for Paul too) one non-techinical problem . . .
Attempt N2 2026-02-19 |Mail list has been broken. . . Or it looks as broken
Attempt N2 2026-02-19 |
Attempt N2 2026-02-19 |(
Attempt N2 2026-02-19 |    I will kill / delete this message later...
Attempt N2 2026-02-19 |) 
Attempt N2 2026-02-19 |
Attempt N2 2026-02-19 |

[pmcjones]

Olaf,

Hi again Paul McJones, thanks for all the work you have invested in the history of Modula-3.

It was my pleasure to be able to carry out this project.

I seem to remember that I still owe you some information, but have forgotten what exactly I wanted to dig out. I'm also not sure if Mike ***@***.***) has sent you the source code archives we still have somewhere.

I haven’t heard from Mike. Last October you said "I’m sure Mike can dig up an old CVS archive somewhere. I’ll also try to remember if and what we merged then."

However, when tidying up some old shelves here in our office, I found an original CM3 Reactor CD from Critical Mass I got from Farshad Nayeri long ago. I've attached a photo just in case you are curious.

The photo did not make it through, but I already have two Reactor CDs on the web site: Distribution CD-ROM. Version 4.0. 6 May 1996 .ISO <https://softwarepreservation.computerhistory.org/modula3/archive/Critical_Mass/Reactor/Reactor_4_0_1996_05_06.iso> Version 4.1. 15 December 1996. .ISO <https://softwarepreservation.computerhistory.org/modula3/archive/Critical_Mass/Reactor/Reactor_4_1_1996_12_15.iso>They are both readable; I also have a full source snapshot from that time: Source code. Critical Mass sources up to 16 December 1996. .ISO <https://softwarepreservation.computerhistory.org/modula3/archive/Critical_Mass/CM-96.iso> Paul

[hosking]

Hi Paul, Yes, thanks for all your work. Regarding UMass GNU Modula-3 sources I am sorry to say that I have looked but not had success in finding them. If Amer Diwan, Eliot Moss, and Rick Hudson (all of whom I think you’ve contacted) don’t have them then I fear that they are lost. Best, Tony From: Paul McJones ***@***.***> Date: Wednesday, February 18, 2026 at 6:35 AM To: modula3/cm3 ***@***.***> Cc: Subscribed ***@***.***> Subject: Re: [modula3/cm3] https://softwarepreservation.computerhistory.org/modula3/ (Discussion #1221) Olaf,

Hi again Paul McJones, thanks for all the work you have invested in the history of Modula-3.

It was my pleasure to be able to carry out this project.

I seem to remember that I still owe you some information, but have forgotten what exactly I wanted to dig out. I'm also not sure if Mike ***@***.***) has sent you the source code archives we still have somewhere.

I haven’t heard from Mike. Last October you said "I’m sure Mike can dig up an old CVS archive somewhere. I’ll also try to remember if and what we merged then."

However, when tidying up some old shelves here in our office, I found an original CM3 Reactor CD from Critical Mass I got from Farshad Nayeri long ago. I've attached a photo just in case you are curious.

The photo did not make it through, but I already have two Reactor CDs on the web site: Distribution CD-ROM. Version 4.0. 6 May 1996 .ISO <https://softwarepreservation.computerhistory.org/modula3/archive/Critical_Mass/Reactor/Reactor_4_0_1996_05_06.iso> Version 4.1. 15 December 1996. .ISO <https://softwarepreservation.computerhistory.org/modula3/archive/Critical_Mass/Reactor/Reactor_4_1_1996_12_15.iso>They are both readable; I also have a full source snapshot from that time: Source code. Critical Mass sources up to 16 December 1996. .ISO <https://softwarepreservation.computerhistory.org/modula3/archive/Critical_Mass/CM-96.iso> Paul — Reply to this email directly, view it on GitHub<#1221 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ABJ2FROPDCLRDL3RRYRX5ID4MNUORAVCNFSM6AAAAACUU6MTDSVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTKOBTHE2TCMI>. You are receiving this because you are subscribed to this thread.

[pmcjones]

Tony, Thanks very much for “closing the loop” on this question. Paul

…

On Feb 17, 2026, at 2:10 PM, Antony Hosking ***@***.***> wrote: Hi Paul, Yes, thanks for all your work. Regarding UMass GNU Modula-3 sources I am sorry to say that I have looked but not had success in finding them. If Amer Diwan, Eliot Moss, and Rick Hudson (all of whom I think you’ve contacted) don’t have them then I fear that they are lost. Best, Tony