Date: Fri, 15 Sep 1995 10:31:55 -0700
From: Joe Catalano
Message-Id: <9509151731.AA13338@library.berkeley.edu>
To: cceseattle@aol.com, doughert@interval.com, jcatalan@library.berkeley.edu,
paulineo@tmn.com, pow@cruzio.com, scot@sbe.csuhayward.edu,
timper@artifact.com
Subject: Tech Committee discussion group
Content-Type: text
Content-Length: 1577
Status: RO
X-Status: A

Hello Everyone!

This is just a note to inaugurate this committee's formation that
we talked about last Saturday in the Redwood Park in Oakland. As
a recap, my sense is that this committee could be used as a forum
to discuss how the Oliveros Foundation can best utilize this nutty
Internet technology to best extend the good work of the foundation.

I think the most efficient way to work the committe is for members
to set up an alias that includes all of our addresses and always respond
to the group using that alias or we could just "reply" to everyone in the
group. This way we have a listserve without having to ask a sysop
to set one up for us. I know I state the obvious, but hey it's still
early in the day with the fog is still heavy on the East Bay Hills.

The thought just occured to me that I should state who is on this
list as we all my not know eachother. So, the members are:

Pauline Oliveros - Artistic Co-Director, POF
Robert Mann - POF Executive Director - POF
Tim Perkis - Composer/Performer and programme, inventor
of electronic instruments.
Scot Gresham-Lancaster - Composer/Performer, electronic
musician and hacker of digital gear of all kinds.
Tom Dougherty - Composer and programmer (Tom what else are you into?, let
the group know).
Greg Powers - Composer/Performer and programmer performing jazz and free
improvisation
Joe Catalano - Composer/performer of really long pieces.

It was great seeing everyone Saturday, especially in a DL context.
(antoher SOTO-Statement of the Obvious, I know).

All the best to everyone!

Joe Catalano

From doughert@interval.com Fri Sep 15 12:00 PDT 1995
Return-Path:
Received: from sbe.csuhayward.edu by s1.csuhayward.edu (5.0/SMI-SVR4)
id AA06176; Fri, 15 Sep 1995 12:00:25 -0700
Received: from fred.interval.com by sbe.csuhayward.edu (5.0/SMI-4.1)
id AA16299; Fri, 15 Sep 95 11:59:01 PDT
Received: from interval.interval.com (interval.interval.com [192.203.7.10]) by fred.interval.com with ESMTP id LAA08319; Fri, 15 Sep 1995 11:59:48 -0700
Received: by interval.interval.com id LAA01969; Fri, 15 Sep 1995 11:59:39 -0700
Message-Id:
Mime-Version: 1.0
Date: Fri, 15 Sep 1995 11:58:49 -0700
To: cceseattle@aol.com, jcatalan@library.berkeley.edu, paulineo@tmn.com,
pow@cruzio.com, scot@sbe.csuhayward.edu, timper@artifact.com
From: doughert@interval.com (Tom Dougherty)
Subject: DL Newsletter Web-isized
Content-Type: text/plain; charset="us-ascii"
Content-Length: 1017
Status: RO
X-Status: A

All -
I realized that I could HTML-isize the Deep Listening Newsletter with
Clarisworks 4.0. So I spent a couple of hours translating it, so that it
could be put on ArtsWire. I'll send the files to Pauline at AOL (since
that's an easy way to deal with enclosures).

Nothing that fancy for the old newsletter, but people will be able to read
it online, and see better versions of some of the artwork (some color).
Maybe more in the next.

As for what I do....
In addition to those things listed by Joe, my background(s) are in
cognitive psychology, drumming and percussion, music perception, sound
design, multimedia, interface, and interaction design. Other interests
include alternative tuning systems, visual and auditory illusions, spastic
rhythms.

Sincerely,

Tom

__________________________________________
T o m D o u g h e r t y , P h D doughert@interval.com
Auditory Interface Design and Research Contractor
Interval Research Corporation
415.949.1929
__________________________________________

From doughert@interval.com Mon Sep 25 10:44 PDT 1995
Return-Path:
Received: from csuhayward.edu (csuh.csuhayward.edu) by s1.csuhayward.edu (5.0/SMI-SVR4)
id AA23352; Mon, 25 Sep 1995 10:44:40 -0700
Received: from csuhayward.csuhayward.edu by csuhayward.edu; (5.65/1.1.8.2/30Jan95-0101PM)
id AA15550; Mon, 25 Sep 1995 10:51:51 -0700
Received: by csuhayward.csuhayward.edu (5.57/Ultrix3.0-C)
id AA08662; Mon, 25 Sep 95 10:47:18 -0700
Received: from fred.interval.com by csuhayward.edu; (5.65/1.1.8.2/30Jan95-0101PM)
id AA15500; Mon, 25 Sep 1995 10:49:28 -0700
Received: from interval.interval.com (interval.interval.com [192.203.7.10]) by fred.interval.com with ESMTP id KAA26765 for ; Mon, 25 Sep 1995 10:42:04 -0700
Received: by interval.interval.com id KAA06531; Mon, 25 Sep 1995 10:42:00 -0700
Message-Id:
Mime-Version: 1.0
Date: Mon, 25 Sep 1995 10:41:05 -0700
To: Scot Gresham-Lancaster
From: doughert@interval.com (Tom Dougherty)
Subject: Re: Contact Sam Ashley...
Content-Type: text/plain; charset="us-ascii"
Content-Length: 1052
Status: RO
X-Status:

>Sam is not presently online and he wants to talk to you about something.
>I came to find out I have neither a snail mail address or phone number
>that is current for you. Could you forward them to me and call Sam at:
>510-428-0578
>Thanx...
>
Scot -
My address is:
317 Ramon Drive, Los Altos, CA 94024
My phone is:
(415) 949-1929

I will call Sam today.

Tom

>x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0x0
>Scot Gresham-Lancaster scot@csuhayward.edu
>MB1525a ph: 510-885-3150
>CSUHayward fax:510-885-3461
>Hayward, CA 94542
>
> http://tesla.csuhayward.edu/~scot >
>
>"Those who sacrifice freedom for security, will have neither" Benjamin Franklin

__________________________________________
T o m D o u g h e r t y , P h D doughert@interval.com
Auditory Interface Design and Research Contractor
Interval Research Corporation
415.949.1929
__________________________________________

From jmax@interport.net Mon Oct 30 12:41 PST 1995
Return-Path:
Received: from sbe.csuhayward.edu by s1.csuhayward.edu (5.0/SMI-SVR4)
id AA12091; Mon, 30 Oct 1995 12:41:37 -0800
Received: from park.interport.net by sbe.csuhayward.edu (5.0/SMI-4.1)
id AA00830; Mon, 30 Oct 95 12:38:38 PST
Received: from [204.74.9.166] (jmax.port.net [204.74.9.166]) by park.interport.net (8.6.11/8.6.11) with SMTP id PAA24173; Mon, 30 Oct 1995 15:40:05 -0500
X-Sender: jmax@interport.net (Unverified)
Message-Id:
Mime-Version: 1.0
Date: Mon, 30 Oct 1995 15:47:22 -0500
To: dgamper@mhv.net (David Gamper), doughert@interval.com
From: jmax@interport.net (John Maxwell Hobbs)
Subject: re: What would it take...?
Cc: cceseattle@aol.com, jcatalan@library.berkeley.edu, paulineo@tmn.com,
pow@cruzio.com, scot@sbe.csuhayward.edu, timper@artifact.com
Content-Type: text/plain; charset="us-ascii"
Content-Length: 6331
Status: RO
X-Status:

At 10:24 AM 10/28/95, David Gamper wrote:
>On 10/12/95, Tom Dougherty wrote:
>>
>>What would it take to do a collaborative audio space on a server somewhere?
>>Pauline had mentioned getting together a system to:
>>* Allow many people to play
>>* Processes the sound with some ambience
>>* Mix all players' sounds together
>>* Allow all to hear the result

Howdy folks,

David Gamper forwarded this to me. I've just had a chance to look it over
and have some thoughts:

* You wouldn't want to do it over the Internet until everyone on the net
is connected with ATM (not likely in the near future) because the
transmission dely on the net is not only substantial, but constantly
variable. There is no way to predict the size of the next packet to be
sent, or what it contains. The server software would have to anticipate
the maximum delay which could be three seconds or more, and impose a hard
delay on top of that that is greater than the overall delay in order to
make the playback consistant - this is what RealAudio does. Try a system
that flows with the delay, like CUSeeMe, and you'll see how that works.
>>
>
>There is hardware (a CODEC) to compress/digitize stereo audio, send over
>phone lines (maybe only ISDN?) and decompress at the other end. I think
>it's still in the multi $K range. Each participating site would need one.
>Server would need many?

* Digital Codecs transmit audio via ISDN and up. The larger the bandwith,
the less compression needed and the faster the transmission speed. The
compression is of the audio signal via PASC or some other form of coding,
and not of the digital stream. As you know, compression of a digital audio
stream is next to useless because of the lack of redundant data. At the
present time, these devices work only point-to-point. The only way to use
them to create a WAN as is proposed, is to set up a large rack of them
feeding to a mixing board. None of these devices do any DSP other than the
PASC compression, so there's no way to control that. A solution would be to
use a phone line to transmit MIDI data to control an automated mixer.

>BTW, the Electronic Cafe has been linking
>audio and video (albiet low rez of both) for some time. David Hykes is
>presenting something at the Kitchen around now involving sending sound to a
>cathedral in France and bringing back its reverb and ambience to the
>Kitchen and elsewhere.

* Low rez is a relative term. The event with David Hykes on the 11th and
12th of November will involve video at 15 fps and 15k stereo digital audio
transmitted at 128k with a one-way transmission delay of approx. 40 ms. at
a transmission rate of 128 kbps. This delay would be reduced to aprox 10ms
if the transmission were done a 256kpbs, which we are considering.
>
>Anyway, I think bandwidth IS the main problem - collecting enough phone
>lines at the server, although player's sites might only need two. Solving
>that, any transmission/codec delays would not be too much of a problem -
>just a part of the process and instrument. The Kitchen minimizes
>transmission delays by calling over and over again until they get a land
>line rather than a satalite link.

* What David is refering to is our Global Drumming Circles. These events
used low-fi, high speed analog connections - in other words - standard
telephone lines. Using an inexpensive part from Radio Shack (the greatest
place in the world!) we were able to interface our sound system an our
phone lines. Telephone bandwidth is 200 - 5,000 hz - not terrible, but
good enough for our purposes in this case: timing information was more
important that harmonic content with improvised drumming. With a land
link, we were right up against the unbeatable delay - 1ms/186 miles - the
speed of light.

Scot Gresham-Lancaster wrote:

>
>In my opinion the only way to get real time audio at this time is with
>dedicated lines, forget using internet because of the inherent timing
>inconsistencies. I suggest that we use a model like that suggested by
>Tim Perkis and Bill Thibault (Openfield project)
>http://tesla.csuhayward.edu/~timper... I can send you the text if the web
>is a problem (do you have lynx there?) if we were to use internet as a
>means of transmission. Another variation on this theme is to remotely
>control identical software instruments at each site. this would greatly cut
>down on the bandwidth problem. Although it would take some of the
>"breath" out of the performance...

* We've been doing this for a long time. Mark Conigilio has written a nice
Mac program call MidiPhone. He's developed a IP/TCP version for the Mac
and is working on a way to set up a limited MIDI server using ISDN - but
not connected to the Internet.

>>From what I understand those performance were done on CuSeeMe or nv and
>therefore didn't have anything over 7K bandwidth with .au compression.
>(ie good for conversation and not much more) I've cc'ed (new verb) this
>to Zvonar since he has been dealing with the Electric Cafe for the last
>few years on several projects.

* The Electronic Cafe has never used CUSeeMe for any interactive purposes.
All of the events are done point-to-point, and don't involve the net. The
Santa Monica site does net "simulcasts" of events via CUSeeMe. In
December, however, the Electronic Cafe @ The Kitchen will be doing the
Electronic Cafe's first interactive CUSeeMe event with Prague. CUSeeMe
will be used only for video however, the audio will be sent over phone
lines.

> On Mon, 30 Oct 1995, Tim Perkis wrote:
>

>> Yes, I think that's a good possibility. Sending ambience from one space to
>> another, like the David Hykes piece mentioned earlier, is cool, but I'm
>> always more interested in the texture introduced by what we're actually
>>>> >> >> doing:
>> what is the "ambience" of the net?
>>

The purpose my idea for the Hykes transmission, (the piece is not designed
for the technology, it's just using it) what that the acoustics of a space
are always going to be better than a computer simulation of it. This is a
reverberation equivalent to the Disklavier and the reverse of Beatlemania
"Not an incredible simulation, but the real thing."

-JMH

John Maxwell Hobbs
Producing Director, The Kitchen http://www.panix.com/kitchen

jmax@interport.net http://www.interport.net/~jmax/

From dgamper@mhv.net Thu Nov 2 01:38 PST 1995
Return-Path:
Received: from csuhayward.edu (csuh.csuhayward.edu) by s1.csuhayward.edu (5.0/SMI-SVR4)
id AA00344; Thu, 2 Nov 1995 01:38:02 -0800
Received: from csbh.mhv.net by csuhayward.edu; (5.65/1.1.8.2/30Jan95-0101PM)
id AA17771; Thu, 2 Nov 1995 01:08:43 -0800
Received: from port106.mhv.net by mhv.net (5.0/SMI-SVR4)
id AB24147; Thu, 2 Nov 1995 04:01:19 -0500
X-Sender: dgamper@pop.mhv.net
Message-Id:
Mime-Version: 1.0
Date: Thu, 2 Nov 1995 05:08:42 -0400
To: Scot Gresham-Lancaster ,
John Maxwell Hobbs
From: dgamper@mhv.net (David Gamper)
Subject: re: What would it take...?
Cc: doughert@interval.com, cceseattle@aol.com, jcatalan@library.berkeley.edu,
paulineo@tmn.com, pow@cruzio.com, scot@sbe.csuhayward.edu,
timper@artifact.com, Richard Zvonar ,
Thibault Bill
Content-Type: text/plain; charset="us-ascii"
Content-Length: 1284
Status: RO
X-Status: D

At 4:21 PM 10/31/95, Scot Gresham-Lancaster wrote:
>If someon can pony up the cash then
>I would love to try out these ideas, otherwise I think we have to settle
>on compromise strategies that don't sacrafice the aesthetic principles
>that are driving our interest in this area...
>

A good point. It's relatively easy to describe a good setup when cost is no
object, but I'm getting stuck in the model. Are there some ideas about how
to not "sacrafice the aesthetic principles" but keep the cost down?

Here's one:
Don't worry about fidelity of the sounds sent to the virtual space. Use as
much bandwidth as economically feasible for them and no more. The server
mixes and distributes the sound contributions AND control data. Everyone
has the same ambient processor (they, at least, are getting cheaper and
more powerful). The control data goes with the sounds, so delays are not as
much of a problem. Delays could be quantized in the processor if desired.
To make sure that everyone is running the same program on the ambient
processor, the MIDI Sysex setup data is broadcast (email binhex) long
before the performance for decoding and downloading into the processor. Of
course, there might be a long battle to decide which processor is the best.
Just a top of head thot.

David