Return to freem_history.ms CVS log

Up to [Coherent Logic Development] / papers / freem_history

Rewording and biblio updates

\"
\" $Id: freem_history.ms,v 1.21 2025/04/28 14:42:52 snw Exp $
\"  History of FreeM
\"
\" Copyright (C) 2025 Serena Willis
\"
\"   $Log: freem_history.ms,v $
\"   Revision 1.21  2025/04/28 14:42:52  snw
\"   Rewording and biblio updates
\"
\"   Revision 1.20  2025/04/28 12:34:14  snw
\"   Update access dates
\"
\"   Revision 1.19  2025/04/24 19:47:33  snw
\"   Improve abstract
\"
\"   Revision 1.18  2025/04/24 18:54:56  snw
\"   Updates
\"
\"   Revision 1.17  2025/04/24 18:46:33  snw
\"   Clean up refs and add ackowledgments
\"
\"   Revision 1.16  2025/04/24 17:07:50  snw
\"   Updates
\"
\"   Revision 1.15  2025/04/24 16:47:58  snw
\"   Reorganize document, add MUMPS language overview
\"
\"   Revision 1.14  2025/04/24 15:33:29  snw
\"   Add table of ISC-acquired M implementations
\"
\"   Revision 1.13  2025/04/24 01:31:13  snw
\"   Add biblio checker
\"
\"   Revision 1.12  2025/04/23 18:59:53  snw
\"   Refine wording
\"
\"   Revision 1.11  2025/04/23 18:58:31  snw
\"   Refine awkward paragraphing
\"
\"   Revision 1.10  2025/04/23 18:44:26  snw
\"   Correct capitalization error in abstract
\"
\"   Revision 1.9  2025/04/23 18:40:00  snw
\"   Improve footnotes
\"
\"   Revision 1.8  2025/04/23 15:12:13  snw
\"   Improve wording
\"
\"   Revision 1.7  2025/04/23 15:05:45  snw
\"   Fix typographical problems
\"
\"   Revision 1.6  2025/04/23 13:32:34  snw
\"   Fix copyright symbol
\"
\"   Revision 1.5  2025/04/23 13:29:37  snw
\"   Add copyright and license, shorten URLs
\"
\"   Revision 1.4  2025/04/22 18:16:06  snw
\"   Remove references to FreeM as a database
\"
\"   Revision 1.3  2025/04/22 17:54:09  snw
\"   Initial draft of FreeM History paper
\"
\"   Revision 1.2  2025/04/22 14:56:22  snw
\"   Add comment header to FreeM History
\"
\"
\" .R1
\" database ../references.bib
\" move-punctuation
\" .R2
.TL
The History of FreeM
.AU
Serena Willis
.AI
Coherent Logic Development
.AB
The history of M, as well as the specifics of FreeM\(emthe implementation maintained by the author for the last decade\(emare extensive and complex topics. Here, we will present an overview of the M language and its history in general, proceeding thence specifically to the historical background and development of FreeM itself, as well as its current status and goals. Though publicly-available information is sparse, efforts have been made to cite as many sources\*{*\*} as possible.
.FS \*{*\*}
Note that all world wide web URLs have been shortened for readability. Following these URLs will automatically redirect the reader to the appropriate, original source. All such URLs were accessed by the author on 27 April 2025, unless otherwise specified.
.FE
.AE
.NH
M LANGUAGE OVERVIEW
.PP
\fIMUMPS\fR\(emnow generally referred to as \fIM\fR to distance it from the infectious disease\(emis an imperative, general-purpose, command-oriented programming language supporting persistent, hierarchical, sparse arrays. It includes high-level support for multi-user, multi-processing applications with flexible concurrency control.
.PP
MUMPS, being an acronym for \fIM\fRassachusetts General Hospital \fIU\fRtility \fIM\fRulti-\fIP\fRrogramming \fIS\fRystem, was developed by Neil Pappalardo, Robert A. Greenes, and Curt Marble of Dr. Octo Barnett's lab at Massachusetts General Hospital in 1966-1967.
.[
greenes1969
.]
Inspired by RAND Corporation's JOSS, and the TELCOMP and STRINGCOMP languages from Bolt, Beranek and Newman, MUMPS' earliest and most prominent use cases were in the field of health informatics.
.[
greenes1970
.]
.[
pendergrass1969
.]
.PP
MUMPS code is divided into program units referred to as \fIroutines\fR, which are roughly analagous to \fImodules\fR or \fIcompilation units\fR in more mainstream languages. Routines are then further divided into labels (or \fItags\fR in MUMPS parlance), which serve as entry points into the routine. Subroutines and functions (the latter known as \fIextrinsic functions\fR) are introduced as tags with an optional, parenthesized \fIformallist\fR, being a list of parameters expected by the subroutine or extrinsic function.
.PP
Perhaps the most unique feature of M, the \fIglobal\fR, is a persistent, hierarchical, sparse array allowing developers to easily implement database features directly in the application's primary development language:
.IP
.DS L
.ft C
ROUTINE ;
  SET ^PEOPLE("Willis, Serena","DOB")="1980-12-01"
  SET ^PEOPLE("Willis, Serena","SEX")="F"
  SET ^PEOPLE("Willis, Serena","COUNTRY")="US" 
.ft
.PP
The above routine would be called with \fIDO ^ROUTINE\fR, resulting in a data structure in fixed storage where the elements (known in M parlance as \fIsubscripts\fR), i.e., \fI"Willis, Serena"\fR, are automatically sorted on insertion, allowing for rapid retrieval. Rough analogues can be found in the \fIPICK\fR operating system, as well as BBN \fIFILECOMP\fR, the latter of which was a direct influence in the early development of M.
.KS
.PP
Early versions of M included \fIMGH MUMPS\fR from Massachusetts General Hospital, and \fIDigital Standard MUMPS\fR from Digital Equipment Corporation. Notably, early commercial versions of M were commercial software, somewhat limiting the language's spread outside of its ecological niche of health informatics.
.TS
center tab(#);
Cb Cb
L L.
Vendor#Implementation
Mass. General Hospital#MGH MUMPS
Digital Equipment Corp.#Digital Standard MUMPS (DSM)
InterSystems Corp.#InterSystems Standard MUMPS (ISM), Open M, Cache, IRIS Data Platform
Data Tree#Data Tree MUMPS (DTM)
Micronetics#Micronetics Standard MUMPS (MSM)
Patterson & Gray#PSM-11, PSM-32, PSM-V
MGlobal International Inc.#CCSM, MacMUMPS, M Global MUMPS, M3 Lite
U.C. Davis#MicroMUMPS
Ray Newman#MUMPS V1
Fourth Watch Software LC#Reference Standard M (RSM)
Kevin O'Kane#Mumps-II, Open Mumps, GPL Mumps
M21 Ltd.#M21
Fidelity National Information Services#GT.M (Greystone Technology MUMPS)
YottaDB#YottaDB
Eugene Karataev#MiniM
.TE
.CD
.SM
\fIFigure 1. Partial list of M implementations.\fR
.NL
.LD
.KE
.NH
EARLY HISTORY OF FREEM
.PP
FreeM was developed in Germany in the mid-1990s by a developer who went by the pseudonym \fIShalom ha-Ashkenaz\fR, whose actual identity remains unknown,
.[
walters99
%P 19
.]
though it is thought by some that they are a dentist who learned C and developed FreeM on their own time. Shalom developed FreeM at a time when InterSystems Corporation (the company that developed the ISM implementation of M) was acquiring the majority of its competitors and absorbing their technology into their \fIOpen M\fR product, which would later become \fIInterSystems Cache\fR.
.[
ibid
%P 18
.]
.KS
.TS
center tab(#);
Cb Cb
L L.
Implementation#Year
DataTree MUMPS (DTM)#1993
Digital Standard MUMPS (DSM)#1995
.[
dsmAcquisition
.]
Micronetics Standard MUMPS (MSM)#1998
.TE
.CD
.SM
\fIFigure 2. M implementations absorbed by InterSystems Corp.\fR
.NL
.LD
.KE
.PP
Shalom wished to provide a community-driven, open-source implementation of M as a bulwark against the growing threat of single-vendor hegemony over the M language. Its design\(emas well as some of the documentation included with the original sources\(emindicate that FreeM was originally targeted to the MS-DOS family of operating systems. It made use of a very limited subset of the C library, and included instructions for renaming the MS-DOS style 8.3 filenames in order to compile under UNIX.
.[
freemREADME
.]
.KS
.PP
At one point in FreeM's early history, Shalom ported FreeM from MS-DOS to SCO UNIX, the UNIX System V Release III-derived descendant of Microsoft XENIX, now known as SCO OpenServer\(ema platform still supported by FreeM today.
.[
portSCO
.]
This port brought support for the \fIscoansi\fR terminal type, including colors and ANSI X3.64 control mnemonics.
.KE
.NH
INITIAL INVOLVEMENT
.PP
The author's mentor in computer programming and UNIX was Lawrence Landis, who involved himself heavily in the M programming language ca. 1991. Mr. Landis promoted the M language to the author from 1991 forward, and first demonstrated FreeM to her in August 1998. In 2010, the author incorporated her company, Coherent Logic Development, learned M, and began doing contract work in M through Mr. Landis's company, Fourth Watch Software.
.PP
Mr. Landis was the owner of FreeM's SourceForge repository
.[
gumpSF
.]
, which had not been touched in a number of years, following Fidelity National Information Services' decision to release GT.M under a free software license. In August 2011, the author downloaded the source code for FreeM and did enough work on it to enable building and running under modern GNU/Linux systems and posted it to the \fImumpster.org\fR forums.
.[
freem014
.]
.PP
In 2014, Mr. Landis gave the author administrator access to the FreeM SourceForge repository and transferred maintainership of the project to her.
.[
gumpSF
.]
.NH
GENERIC UNIVERSAL M PROJECT
.PP
The \fIGeneric Universal M Project\fR was conceived by Richard F. Walters, a professor from U.C. Davis. The GUMP, following the rising popularity of object-oriented programming in the 1990s, was intended to be a toolkit allowing M implementations to be built from discrete components with a well-defined and well-specified public interface among these components. These components included the global handler (supplying the functionality of persistent global storage), and the interpreter/compiler (responsible for implementing M language commands). The components would have been able to communicate over a network, or in-process on the same host, enabling distributed computing functionality.
.[
c2m
.]
.PP
Although the specification for the GUM interface to global handlers attained a reasonably well-specified level of completeness,
.[
gumapi
.]
and Lawrence Landis and others developed a mostly-complete implementation of a GUM global handler,
.[
gumpsrc
.]
none of the other envisioned components were ever completed, and specifically, the interpreter component was missing.
.NH
FREEM DONATED TO MUG DEUTSCHLAND
.PP
In July of 1998, Shalom ha-Ashkenaz donated the FreeM source code (then known as FreeMUMPS) to the M User's Group-Deutschland (MUG-D), hoping the community would take the nascent implementation from its infancy through to a state of production-ready completeness and robustness. Shalom also placed a few conditions on his gift: a public release could not be made until a substantial set of milestones were reached. Per his conditions, the FreeMUMPS project must:
.[
freemREADME
.]
.IP \(bu 2
Implement the entirety of \fIANSI X11.1-1995\fR
.IP \(bu 2
Use Structured System Variables instead of \fIVIEW\fR commands and \fI$VIEW\fR functions
.IP \(bu 2
Raise the string size limits
.IP \(bu 2
Implement MWAPI, OMI, X11 bindings, and GKS bindings
.IP \(bu 2
Be substantially free of major bugs
.PP
Although MUG-D readily accepted the contribution of FreeMUMPS, the organization itself lacked the manpower and expertise to complete the implementation. Just as it is now, the intersection of M community members who know enough of the M language and C language to work on a project this ambitious was quite small.
.NH
MERGING GUMP AND FREEM
.PP
Very shortly after the contribution of FreeMUMPS to MUG-D, Richard F. Walters and a small team of developers and administrative staff who had been working on the GUMP assumed maintainership of the FreeMUMPS source code, with Lawrence Landis managing the development efforts.
.[
shortWalters99
%P 19
.]
This included representatives from the \fIM Technology Association\fR (an M vendor association having several foreign branches), the \fIM Development Committee\fR (the M standards organization hosting the ANSI/ISO standards for the M language, then sponsored by the M Technology Association), and others.
.[
freemContributors
.]
The goals of this team were to:
.IP \(bu 2
Meet Shalom's requirements for a public release of FreeMUMPS
.IP \(bu 2
Convert FreeMUMPS into the first interpreter component of the GUMP
.PP
During this period, Ronald L. Fox of Diagnostic Laboratory Services in Honolulu, HI (who passed in 2010)
.[
ronFoxGrave
.]
ported FreeMUMPS from SCO UNIX to Red Hat 5 and glibc-6.
.[
ronFoxPort
.]
Steve "Saintly" Zeck of the U.C. Davis Veterinary Medical Teaching Hospital
.[
saintlyBio
.]
also attempted to rewrite the symbol table code to lift string size limits,
.[
saintlySymtab
.]
David Whitten enhanced some of the implementation-specific extensions, and Lawrence Landis integrated Zeck's symbol table work.
.PP
In FreeM 0.1.0, the name of the implementation was changed from FreeMUMPS to Public Standard M, and again to Free Standard MUMPS and then FreeM when it was discovered leading up to the FreeM 0.2.0 release that the PSM acronym was already in use for Patterson & Gray's
.[
m_implementations
.]
M implementation.
.[
Changes.GUM
.]
Dr. Walters also received the implementation ID of 49 from then secretary of the M Development Committee, Don Piccone.
.[
ibid
.]
.PP
One of the contributors to FreeM at this stage\(emprimarily in the area of M vendor routines\(emwas Axel Trocha, who would later maintain a private fork of FreeM.
.[
trochaFork
.]
.NH
GT.M RELEASED AS FREE SOFTWARE
.PP
GT.M, an acronym for \fIGreystone Technology MUMPS\fR, is an M implementation that was released by Greystone Technology in 1986. Greystone was later acquired by Sanchez Computer Associates, which was in turn acquired by Fidelity National Information Services.
.[
fisSanchezAcquisition
.]
.KS
.PP
When GT.M was released under the GNU General Public License in 2000,
.[
gtmRelease
.]
it seemed to many to obviate the \fIraison d'etre\fR for FreeM, as GT.M was a well-established, robust, and high-performance M implementation with which FreeM could not then compete. Unfortunately, at this time, the GUMP and FreeM projects lost all of their momentum, and new development along these lines rapidly ceased. The final GUMP team release of FreeM was 0.5.0.
.[
freem050
.]
However, Axel Trocha's private port would continue to be developed for some years.
.KE
.NH
FORKED BY AXEL TROCHA
.PP
When the free software release of GT.M stalled the GUMP team's progress on the primary branch of development, Axel Trocha, an aforementioned contributor, continued development on the FreeM source code. Trocha added many interesting features to the FreeM codebase, including:
.IP \(bu 2
A native port to Microsoft Windows
.IP \(bu 2
Compiling FreeM as an Apache web server module, allowing FreeM to be used easily for web development
.IP \(bu 2
The ability to output HTML code in a heredoc-style format, with any line of code beginning with a left angle bracket being interpreted as HTML with support for interpolated M locals and globals
.IP \(bu 2
Extensions allowing FreeM to be used as a command-line shell, along the lines of UNIX \fIbash\fR, Windows \fIcmd.exe\fR, etc.
.PP
Trocha also maintains ownership of the \fIfreem.net\fR Internet domain,
.[
trochaDomain
.]
and continued issuing public releases of his FreeM port on that site until sometime after 2004, at which point this fork was made entirely private. Currently, freem.net is a blank page. However, Trocha's fork of FreeM continues to the present
.[
trochaEmail
.]
as the back-end storage engine and programming environment for the \fIElven Runes\fR website.
.[
elvenRunes
.]
The author has communicated with Mr. Trocha on occasion, and though he is supportive of the author's efforts, has chosen to remain in the background.
.NH
RESUMING PRIMARY DEVELOPMENT
.PP
In 2011, the author downloaded the FreeM source code from the GUM Project's SourceForge repository\(emdormant since 2000\(emand updated it just enough that it would compile and run on modern GNU/Linux systems. The author also quickly updated FreeM to support terminal sizes larger than 80x24.
.NH
TAKING MAINTAINERSHIP
.PP
In 2014, Lawrence Landis transferred administrative access of the GUMP repository, conferring maintainership of the primary branch of FreeM development to the author. Since then, many features have been added and many bugs corrected, including:
.IP \(bu 2
Adding support for proper namespaces, configured through a \fIfreem.conf\fR file, which standardizes routine and global storage locations
.IP \(bu 2
Adding support for Structured System Variables
.IP \(bu 2
Adding support for the asynchronous event specification from MDC Type A proposal \fIX11/1998-28\fR
.[
x119828
.]
.IP \(bu 2
Adding support for constants via the \fICONST\fR keyword
.IP \(bu 2
Adding a \fIWITH\fR command allowing the specification of an implicit prefix to all subsequent variable references
.IP \(bu 2
Adding a runtime \fIWATCH\fR command, tracking changes to specified local or global variables
.IP \(bu 2
Adding an \fIASSERT\fR command, which will fail with an error message if the following expression evaluates \fIfalse\fR
.IP \(bu 2
Removing the Steve Zeck symbol table implementation\(emwhich was unreliable\(emand reverting to the original implementation
.IP \(bu 2
Adding support for the GNU \fIreadline\fR library, with persistent command line history and editing
.IP \(bu 2
Adding REPL-like functionality (in direct mode, any M expression beginning with a number will be prepended with an implicit \fIWRITE\fR)
.IP \(bu 2
Adding transaction processing
.IP \(bu 2
Adding \fIKVALUE\fR and \fIKSUBSCRIPTS\fR
.IP \(bu 2
Preparing to support the \fIM Windowing API\fR (MWAPI)
.IP \(bu 2
Adding the \fIfmadm\fR command-line utility, for system administration functions
.IP \(bu 2
Adding support for after-image journaling and forward recovery of globals
.KS
.IP \(bu 2
Writing a \fItexinfo\fR manual, from which the HTML manual is derived
.IP \(bu 2
Porting to Solaris/SPARC, Solaris/x86, Linux/s390x, Linux/armv6l, Linux/armv7l, SCO OpenServer 5.0.7, Tru64 UNIX/alpha, AIX/ppc, Mac OS X/x86, GNU HURD, Cygwin, NetBSD, FreeBSD, OpenBSD, and WSL1/2

In addition, the FreeM web site
.[
freemWebsite
.]
was created, to distribute downloads and documentation.
.KE
.NH
FUTURE
.PP
FreeM is envisioned as a client-oriented desktop M implementation, for developing graphical user interfaces that will run on mobile and desktop devices.
.PP
The author also intends to adopt the original vision of the GUMP team, dividing FreeM's functionality into discrete components having a well-specified public interface, with the ability to run in distributed computing environments over a network.
.PP
FreeM's mission is to advance the state-of-the-art in M implementations, and push the evolution of the language forward. Maintaining portability to as many vintage and modern UNIX systems as possible is held as a high priority, while portability of M routines and MDC standards compliance will be maintained through the use of the new \fI$ZDIALECT\fR intrinsic special variable to ensure that such compliance does not conflict with the primary goal of elegantly advancing the state-of-the-art and finding new audiences for the concepts originated by Neil Pappalardo and Octo Barnett in 1966.
.PP
The FreeM project is also strongly committed to free software principles, and is firmly aligned with the goals of the GNU Project and the Free Software Foundation, believing that the ethical concerns surrounding proprietary software are at least as important as the practical concerns espoused by the Open Source movement.
.[
whyFreeSoftware
.]
.NH
CONCLUSION
.PP
FreeM has seen a colorful and turbulent history, touched by many capable hands. Though public development of any strain of the implementation was dormant for nearly a decade, fourteen years of almost continuous development have passed since the project resumed in 2011, and a decade since official maintainership passed in 2014, and great progress has been made.
.PP
FreeM\(emas is the case for all M projects\(empresses forward in a period where the future of the M programming language is uncertain. M Development Committee efforts concurrent with FreeM development have been sporadic and have missed many milestones, the community's most prominent members are aging, and many organizations have migrated from large M applications to what is perceived as more "modern" replacements.
.PP
It is the opinion of the author that the tight integration of an expressive and dynamic language with a robust and performant persistent storage engine makes M a natural candidate for many new and general applications. Modern application development is plagued by overwhelming bars to entry, requiring mastery of many languages and database management systems and the bulky interfaces connecting them.
.PP
Though not without significant warts accreted over the years, M has no such overhead. Thus, FreeM seeks to press its philosophical advantages by mitigating the language's accumulated cruft, and adding clean interfaces addressing the needs of today. With these goals in mind, development proceeds apace.
.KS
\" .NH
\" ACKNOWLEDGMENTS
\" .PP
\" Certainly, the intellects responsible for giving us M are too numerous to mention by name, but their importance cannot be overstated. The author would like to especially credit Lawrence Landis for his early and innumerable contributions to her development, both generally as a practitioner of software development, and specifically for introducing her to the M programming language, and inculcating in her a love of exploration and innovation.
\" .PP
\" Thanks are due also to David Wicksell, for countless hours of collaboration on FreeM spanning more than a decade, and for providing ideas and inspiration.
\" .PP
\" I would also like to thank Ken McGlothlen of Vista Expertise Network for his efforts in editing multiple drafts of this paper.
\" .PP
\" Finally, to my wife, Miriam\(ema brilliant technologist in her own right\(emand daughters, without whose inestimable patience and understanding as the author buries herself in countless passion projects, such projects would not be possible.
\" .KE
.KS
.NH
COPYRIGHT AND LICENSE
.LP
This document is Copyright \[co] 2025 Serena Willis
.LP
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License,
.[
gfdl
.]
Version 1.3 or any later version published by the Free Software Foundation; with no Invariant Sections, with no Front-Cover texts, and with no Back-Cover Texts. 
.LP
\fI$Id: freem_history.ms,v 1.21 2025/04/28 14:42:52 snw Exp $\fR
.KE