hal (01/17/83)
Several weeks ago I sent out a request for information about music software
on behalf of the Cornell music department. They are trying to find a way
to automate the production of music theses, particularly in musicology.
I've finally gotten around to sorting out the replies (over 40 of them),
and here's what I found out. Many thanks to all who took time to reply.
Hal Perkins
uucp: {decvax|vax135|...}!cornell!hal
arpa: hal@cornell
bitnet: crnlcs%hal
The most widely known music software in Unix land appears to be Don Byrd's
SMUT program, which was used to print the music in Hofstadter's "Godel,
Escher, Bach." Lots of folks mentioned this book and the music in it; one
person actually sent a mailing address.
----------------------------------------------------------------------------
>From vax135!floyd!harpo!ihnp4!ixn5c!inuxc!pur-ee!iuvax!isrnix!nick Sat Nov 20 00:16:28 1982
Date: 19 Nov 82 12:49:58 EST (Fri)
Status: RO
Don Byrd
School of Music
Indiana University
Bloomington, In 47405
has several routines that do most of what you want.
He is not on the net, but a letter to him would work.
The PDP and VAX are a separate system from the CDC 6600 that the
music transcription programs that he has written on.
Doug Hofstadter in Godel Escher Bach uses some of Don's
routines to write the fugues in the book.
If you have trouble try calling me
Nick Mullins
Sociology
IU
(812) 335-7795 or 335-5127
-----------------------------------------------------------------------
There appear to be a number of systems that can at least print music,
if not edit it.
-----------------------------------------------------------------------
>From allegra!rochester!Root.Boy Thu Dec 9 03:19:22 1982
Date: 8 Dec 1982 23:29-EST
From: Lee.Moore
Subject: music notation for computers
To: allegra!cornell!hal
Message-Id: <82/12/08 2329.583@Super-Vaxen>
Origin: Super-Vaxen
Status: RO
Below is a reply to your news item of sometime ago. I am
typing it in for a friend (Craig Harris). Any further correspondence
should be sent to him: rochester!esm.
Just to stick my own two cents in, I have seen Leyland Smith's scoring
program and it can be impressive. It's bad features are it's written
in Fortran and PDP-10 assembly -- and very poorly at that. The only
thing that keeps it running is constant maintenance. It used a III
scope at SAIL. I don't know if he has converted to work with anything
else. He seems to have a paper about it at every computer music
conference.
--lee
--------------------------------
Yes, there seems to be a lot of work being done in transcribing
music using computers. I'm not sure how successful they have gotten,
but there are some commercially available items, as will as a fair
amount of research being don. Below is a list that might be of
interest to your inquirere.
----------------------------------
Mockingbird- Palo Alto Research Center (PARC), Xerox
-more for composers than researchers
Gregory's Scribe- Interactive graphics for music notation before 1600
DARMS(Digital Alternate Representation of Music Scores)-
-symbolic/graphic-not notes
Harry B. Lincoln- SUNY Binghampton- Encoding, decoding, and storing
melodies for a large database (Renaissance Polyphony)-perhaps
he has made progress since 1977
-----------------------------------
There are two commercially available products, and I'm looking
into those- I don't know them off-hand, but will let you know.
>From kevin Sun Nov 21 00:27:03 1982
To: hal
Subject: music printing software.
Status: RO
Leland Smith at Stanford has written some music printing software that produces
better output than any other I've seen. It will even do styles like square
notes and four-line staves (useful for music historians). The input format
must be the world's worst, though.
At one time I had a manual for the system. If you're interested I'll look
through my bookcase and see if I still have it.
Let me know if anything useful comes from your request.
Kevin Karplus
>From vax135!ariel!orion!lime!we13!otuxa!ll1!ihldt!ihnp4!harpo!floyd!cmcl2!philabs!sdcsvax!sdcattb!sdcarl!lin Fri Nov 26 14:56:49 1982
Subject: Computer graphics for musical notation
Newsgroups: net.music
Status: RO
Professor Leland Smith of Stanford has developed a music notation
system writen in FORTRAN. It has some very nice features -
after the user enters the notes to be played, the program will
rearrange the notes with the correct spacing. This is not a
commercial system; I do not know whether Smith is giving the
programs to other universities for non-profit purposes.
There was also a European system announced a couple of years ago
in the Computer Music Journal. I don't remember anything about
it.
>From vax135!allegra!sbcs!lw Thu Dec 2 06:31:10 1982
Date: 2 Dec 1982 4:35-EST
From: Larry Wittie <lw@suny-sbcs@UDel-Relay>
Subject: music scoring program(s)
To: allegra!vax135!cornell!hal
Status: RO
David Gomberg's dissertation (Wash. U, St. Louis about 1976)
was a music scoring system. There was some talk of making
a commercial product around it. He probably knows of other
systems around the country. He currently works at
Mitre in McLean Va and is listed in the NW corner of DC phone books
not quite as far out as Rockville MD.
>From floyd!stan Wed Nov 17 12:46:13 1982
To: vax135!cornell!hal
Subject: music printing software
Status: RO
At U of I, I knew of some people at the Plato project who drew music
on a plotter; it looked real nice. A guy named Joe Cychosz was involved,
but he is with CDC now. They also had a project that drove a systhesizer
from a programmable Plato terminal, so they may have had editting software.
If you have Plato access, their system name is cerl. If you don't,
Plato's mailing address is:
252 ERL
University of Illinois
<street address goes here>
Urbana, Ill. 61801
Stan King, BTL Wh.
>From vax135!floyd!harpo!decvax!ittvax!dcdwest!wsp Thu Nov 18 02:02:19 1982
Date: Wed Nov 17 18:22:24 1982
To: ittvax!decvax!harpo!floyd!vax135!cornell!hal
Subject: Music Plotters
Cc: wsp
Status: RO
When I was at the University of Michigan, I saw a program which was running
on the Amdahl 470 that plotted music on the CalComp plotter and on a Tektronix
screen. A person there whom you might contact is:
Bernard Galler
Associate Director
Computing Center
University of Michigan
Ann Arbor, MI 48109
He is likely to be able to help you. Mention my name if you would
like.
Peter Benson
ittdcd-west
(619) 578-3080
>From decvax!pur-ee!uiucdcs!kolstad Sun Dec 12 00:54:05 1982
Date: Sat Dec 11 17:25:28 1982
To: pur-ee!decvax!cornell!hal
Subject: music printing
Status: R
If I forgot to tell you: Lippold Haken at 217-333-0766 does all kinds of
this stuff and would be delighted to talk to you. He has tens of thousands
of lines of code for such things.
>From vax135!floyd!harpo!decvax!yale-comix!hawley Thu Nov 18 02:03:08 1982
Date: Wed Nov 17 19:07:34 1982
To: decvax!harpo!floyd!vax135!cornell!hal
Subject: music printing systems...
Cc: peierls
Status: RO
Hey, let me know anything you hear about this.
I've been thinking about working up a music printing system here,
very flexible, with a device independent low-level graphics library
that I can hook onto our spiffy laserprinter.
I haven't found software yet, but just yesterday I read through a
reference manual on DARMS (digital to analog representation of musical symbols)
This was the 1976 manual, written by ... (name escapes me) from Queens college.
Darms is tedious, but very complete and slightly mnemonic.
It allows you to embed text and stuff like that in the score.
You can do really picayune things, like adjust stems.
Ah yes -- Raymond Erickson was the author of this book.
Surely there must be a system around which uses this.
Also, Bill Buxton at UofT can probably help you out.
- Mike Hawley
decvax!yale-comoix!hawley
-------------------------------------------------------------------
As far as interactive music editors, the two best leads were the
Mockingbird system at Xerox PARC and MIT's Music-11
-------------------------------------------------------------------
>From vax135!floyd!harpo!decvax!microsof!uw-beaver!uw-june!suest Fri Nov 19 06:43:00 1982
Date: Thu Nov 18 12:06:57 1982
To: uw-beave!microsof!decvax!harpo!floyd!vax135!cornell!hal
Status: RO
I'd be interested in hearing about any music editing systems you find. The only
thing I've seen is one called "Mockingbird", a music editor which runs on a
XEROX Dorado. It was written by John Maxwell as part of his Master's at MIT.
I've seen a video tape of a demo, and it looks pretty sharp (no pun intended).
susan st. john
uw-beaver!suest
(ARPA) suest@washington
>From vax135!eagle!mit-vax!mp Mon Nov 22 01:17:35 1982
To: eagle!vax135!cornell!hal
Subject: music editor
Status: RO
The Experimental Music Studio here has a music editor that
runs on an Imlac and probably other terminals I think
it's being marketed as MUSIC-11. Also, there's a music editor
for the Lisp Machine, but it's probably not very transportable
to UNIX environments unless you know Lisp.
--------------------------------------------------------------------
Eastman seems to be doing something with either electronic music or
music software or both, I'm not really sure....
--------------------------------------------------------------------
>From vax135!floyd!cmcl2!philabs!sdcsvax!sdccsu3!sdcvax!mike (Michael Williams) Wed Dec 15 08:42:57 1982
Subject: "Re: Computer Music Conference in Italy"
Newsgroups: net.music
From what I've heard (having been there last year), I
understand the '83 CM Conf. will be at Eastman School
of Music (part of the University of Rochester), in
Rochester, NY. They are running Music 11 with a special
preprocessor, which I understand will also be available
for the CARL C-Music as well, called Score 11, written
by Alec Brinkman. They run on a pitifully slow 11-34
with a CDC 300 Mb drive. I have yet to hear any
consistently good compositions or composers of Computer
Music, however Paul Lansky (on the faculty at U. of
Baltimore ?) has written at least two really good pieces
concentrating on LPC techniques.
Mike Williams
Mail drop 72-02
System Development Corp.
2500 Colorado Ave.
Santa Monica, CA. 90406
(213) 820-4111 x5283
UUCP: randvax!sdcvax!mike
burdvax!sdcvax!mike
ucla-vax!sdcvax!mike
sdccsu3!sdcvax!mike
-------------------------------------------------------------------
I'm not sure what's happening out at UCSD either, but I got several
notes suggesting I check there.
-------------------------------------------------------------------
>From vax135!floyd!cmcl2!philabs!dal Thu Nov 18 23:56:10 1982
Date: Thu Nov 18 14:17:49 1982
To: cmcl2!floyd!vax135!cornell!hal
Subject: music editing/printing software wanted
Status: RO
Hal,
In response to your usenet query....
There is a site on the net that has lots of music software. It's called
sdcarl (stands for ucSD Computer Audio Research Lab). There is an article
about them in "Computer Music Journal," vol 6, #1, Spring 1982. The article
says that they will distribute a "snapshot" of their software to interested
parties.
I found out (from a friend at ucsd) that the person to talk to is
sdcarl!rusty. I sent a request to him (it?) but did not receive a reply. As
I am not in the habit of bothering people, I haven't followed up, but if you
get something from them, I would be interested.
To get to sdcarl, you can go through sdcsvax, i.e., sdcsvax!sdcarl!rusty. If
you don't have a better route to sdcsvax, you can go through us, i.e.,
philabs!sdcsvax!sdcarl!rusty
The article does not go into fantastic detail about their music editing
software, but they do a lot of sophisticated stuff and I would be surprised
if it didn't fill your bill.
By the way, it seems odd that your music department doesn't know about
"Computer Music Journal" -- it is published by MIT Press.
Good luck in your search, and, as I said before, if you do get something good
in the way of a music editor, let me know. Thanks.
Dan Lorenzini Philips Laboratories (914) 945-6236 philabs!dal
>From rabbit!jj Wed Nov 17 13:05:07 1982
To: cornell!hal
Subj: Music writing/notation software
Status: RO
You might contact
sdcarl!frm for some idea of what he used to have. (it was quite complex,
as I understand, I dunno what it ran under.)
sdcarl!frm is F(rank) Richard (Dick) Moore, PhD, etc.etc and he heads
the computer music group at UCSD, or at least runs it, if not heads it.
You will have to find your own path.
>From vax135!floyd!cmcl2!philabs!sdcsvax!logo Sat Nov 20 20:58:30 1982
Date: 18 November 1982 1809-PST (Thursday)
Reply-To: logo
To: philabs!cmcl2!floyd!vax135!cornell!hal
Subject: score creator
Status: RO
resources...
Center for Music Experiment allegra!sdcarl or ucbvax!sdcsvax!sdcarl
probably should send note to (hmmm) rusty and ask him to show it to
people. frm would KNOW but he is often REALLY BUSY. try frm and if
no response try rusty.
Computer Music Journal published by MIT Press
at least one article within the last year
many of the names and places mentioned in the CMJ would be worth
contacting. try at least IRCAM in france.
A paper delivered on Mon or Tues at the most recent Audio Engineering
Society convention in Los Angeles. not a very sophisticated system though.
There are adds in Byte and other personal computer magazines for this sort
of thing. there might be a sophisticated sleeper in the lot.
I am a computer consultant and digital audio and synthesis is a hobby of
mine. If I might be of some profesional help, let me know.
David (Reisner)
uucp : ...!ucbvax!sdcsvax!logo
arpanet : sdcsvax!logo@nosc
------------------------------------------------------------------------
A couple of people wrote about music machines that can store and play
music.
------------------------------------------------------------------------
>From vax135!floyd!harpo!decvax!minow Tue Dec 28 19:09:08 1982
Subject: Re: All I Want is Music, Music, Music - (nf)
Newsgroups: net.wanted
Status: RO
All you people out there who are looking for music editors and
composition systems should hunt down information on the Maclavier
(I hope that's spelled right), designed by Dave Maclay in Toronto.
Dave is a musician/composer, successful enough to have bought a
VAX 780 for his own synthesizer. When I saw the system about 3
years ago, he demonstrated it by composing a small piece of music
in about 1/2 hour, laying it down track by track; melody, counterpoint,
rhythm, and all. Then, the software took the notes (as recorded from
the synthesizer keyboard) and printed it out in standard music
notation.
The system looks good, sounds good, and would seem to be ideal for
both composers and researchers. You could probably find a reference
or ad in the Computer Music Journal, published by MIT press.
Martin Minow
decvax!minow
>From vax135!floyd!harpo!allegra!deb Thu Nov 18 01:57:56 1982
Date: 17 Nov 82 17:18:58 EST (Wed)
To: harpo!floyd!vax135!cornell!hal
Subject: Music
Status: RO
I don't know about editing, but a friend of mine wrote a music
processor. Give it music, it plays it: up to 8 voices, any
speed range (MM 88 - 214) about six octaves. If you need
a language (syntax) for describing the music, his syntax is very good.
That's all I can offer.
David Baraff
Murray Hill Bell Labs
..mhtsa!allegra!deb
------------------------------------------------------------------------
But no one knew of anything nearly as sophisticated as my music friends
are looking for. A couple of people suggested I contact Bill Buxton at
Toronto. Before I had a chance to send a note to him, he sent one to
me. I'll close with it, because I think he has the most realistic
perspective on the amount of work it will take train a computer to edit
and print music well enough for scholarly work.
------------------------------------------------------------------------
>From vax135!floyd!harpo!decvax!utzoo!utcsrgv!willy Thu Nov 25 06:34:29 1982
Date: Tue Nov 23 11:30:24 1982
To: utzoo!decvax!harpo!floyd!vax135!cornell!hal
Subject: music editing
Status: RO
Regarding your request for information on music editing, the only thing
that one can say about software to meet your music department's request is that
it is impossible at this point in time. There are several toy music editors
available for personal computers. They are totally incapable of the full
task, or even a minute portion of it. There are a few powerful score
printing packages, but they are not available, complete, or especially
usable. The only civilized interactive system (civilized in that it
even comes close to, for example, a word processor/typesetting system) is
MOCKINGBIRD developed by Zevero Ornstein and John Maxwell at Xerox PARC.
It requires a dorado to run, however, and still misses a huge part
of your specifications. Leland Smith at Stanford has a remarkable
system which runs on their Funglee (KL-10 look-alike), but it also
is not generally available, and has a terrible user interface (by
modern standards). Don Byrd at the University of Indiana also has
an excellent system (for example, all music in Godel Escher & Bach were
typeset using that system). However, it is a non interactive FORTRAN
program that happens to give graphics output. There exists no
comprehensive interactive score editor, for the very good reason of
the incredible difficulty of the task (both in terms of the semantics
of the transactions, the complexity of music typesetting rules, AI
issues, and the purely graphical problems). There have been, since
1965 about 20 PhD theses on the subject. One reply to your mail
suggested that such systems were common, in commercial use, and
commercially available. Nonsense. It is significant that the
respondent did not give one example along with his helpful
(wishful?) comments. It is true that perhaps 70% of music
published in the US is typeset using computer graphics. This
is by a company Music Reproductics in NYC. However, note that
that music is all popular piano renditions of pop songs. Hence,
the software needs only deal with a minute portion of the
posible problems. Also, note that all curved lines (slurs
etc) plus many other parts of the graphics are done by hand
still, even on this system.
One final system of interest, by the way, is NEDIT developed
at the experimental music studio at MIT. It is a truly interactive
editor for score notation, but again, is not generally available,
and misses many of the features you requested.
Without wanting to rain on the music departments parade, there is
nothing that meets what they want, nor will there be for a long time.
Get them a subscription to the Computer Music Journal (MIT Press),
go through back issues in order to get some feeling for what has been
done, and if they still want something, then they are going to have
to revise their wish list to something about 10% of the current one,
and then they might get somewhere. By the way, with respect to
interaction, any supposedly interactive graphics score editor which
attempts to run on a TEK storage tube is a contradiction in terms.
There can be no interaction that deserves the name. The task requires,
from a purely graphics point of view either (a) a high-resolution
(min 1k x 1k) bit mapped display, (b) a medium resolution raster
display capable of displaying anti-aliased characters, or (c) a
high quality caligraphic display, such as E&S or 3 Rivers (capable of
30-50,000 vectors in 30th of a sec).
Some of these statements may seem dogmatic, but that grew over several
years of working on the problems, and continually watching people
embark on doomed projects, or announcing products which simply did
not work. Feel free to get in touch if you have any questions.
Bill Buxton
CSRG
University of Toronto