[Indigo]<Voice>Stewart>plansvp.bravo!1

By this Fall we plan to have the Alto I Etherphone I prototypes (2 to 5 of them) able to talk to each other and to standard phone lines with the aid of a first Etherphone Server. We plan to have essentially pfrozen the Ethernet voice transmission protocol and to have collected information about (and perhaps measured) Ethernet performance for voice traffic. We are already collecting information we need to start the Etherphone II design.

Dan Swinehart, Larry Stewart, and Severo Ornstein are the primary participants. They are spending roughly 70% of their time on the Voice project. In addition John Ousterhout and Susan Owicki are working on the project - John, one day a week, and Susan, one day a week for now and more after mid-October.

Specific roles are as follows:

Dan:Define protocols; work on Etherphone Server program
Larry:Work on Etherphone hardware and software
Severo:Manage things and work on Etherphone hardware with Larry
John:Work on Voice File Server
Susan:Work on Etherphone Server design with Dan and File
Server design with John

Dan and Susan will develop the Etherphone Server code on Dan’s Dolphin. If we can’t find another machine by the time we’re ready to start incorporating the Etherphone Server with the Etherphones, we’ll use Dan’s Dolphin as the Server - and get him an Alto II.

John will develop the Voice File Server code on a Dorado. (He has special sign-up dispensation for his one day a week here). Larry Stewart has a Dolphin with the understanding that if, by the time we are ready to start incorporating the Voice File Server with the other machines, we can’t find another one for the server, we will use his machine - and replace it with an Alto II.

We have ordered a number of parts for the Etherphone II and are beginning to try out some small assemblages to become familiar with how the parts work. We hope to have a working breadboard by about the end of the year. Most of our hardware work will consist of plugging together small numbers of reasonably high level standard parts on prototype boards. A lab bench or table will do for this. Also we’ll need some small power supplies and a scope but we don’t forsee need for substantial lab space. On the other hand, during development, we will want to assemble all of the elements of the distributed system in one room, to make debugging feasible. When fullblown this could include the following items of equipment:

1 or 2 Alto based Etherphone I’s
1 or 2 prototype Etherphone II’s
1 Dolphin Voice File Server
1 T-80 for Voice File Server
1 Dolphin Etherphone Server
1 control ("Midas")* Alto for Etherphone II’s
1 3MB to 1.5 MB Ethernet Alto Gateway (could be elsewhere)
Scopes, tables, etc.

Not a trivial pile of stuff. We plan to use the Nursery. It’s about the right size and near the lab. (Our use fits the mold of the Nursery’s original purpose). Over the next few weeks it will be cleared of other occupants and we will start to move in - at first with the Alto I Etherphone 0’s (presently in Dan and Larry’s offices) together with our hardware breadboarding. Gradually the other stuff will assemble there. We hope to have a rudimentary combined system working next spring. By next summer we expect to have the Etherphone II design solidified enough to order enough units to outfit everyone in CSL with an Etherphone. We have allocated $50K in the 1982 CSL budget for this purpose. By the end of 1982 we hope to have a working backbone system in place in CSL that will handle telephone calls in standard ways, permit simple storage and retrieval of voice messages, and generally put us in a position to start experimenting with more estoeric uses of the voice handling facilities - e.g. integration with Laurel (or derivative).