Author: at INTERNET-MAIL Date: 7/27/97 3:24 PM Priority: Normal TO: rcbraley@fedex.com at INTERNET-MAIL TO: bheile@bbn.com at INTERNET-MAIL TO: mmunson@mobilnet.gte.com at INTERNET-MAIL TO: scase@milkyway.flexipc.com at INTERNET-MAIL TO: jcarroll@milkyway.flexipc.com at INTERNET-MAIL TO: nlemmon@fedex.com at INTERNET-MAIL TO: Dennis Martinez at REMOTE_PO TO: Ian Gifford (Marketing) at PB_HGrail1 TO: rehmi@media.mit.edu at INTERNET-MAIL TO: tia.runner@a90.mort.com at INTERNET-MAIL TO: rhodes@media.mit.edu at INTERNET-MAIL TO: rogerh@xetron.com at INTERNET-MAIL TO: larryo@xetron.com at INTERNET-MAIL TO: gallop@symbol.com at INTERNET-MAIL TO: dvorakj@plnt004.comm.mot.com at INTERNET-MAIL TO: crg005@email.mot.com at INTERNET-MAIL Subject: wearable computing ieee standards meeting notes ------------------------------- Message Contents ------------------------------- Here are the notes I took from the PASC Wearables Standards Meeting, July 16, 1997. If I got anything badly wrong, please send corrections. -- Bradley Rhodes, MIT Media Lab ----------------------------------------------------------------- From: dick braley Subject: "Wearables" Standards Meeting in Nashua Following is the agenda for the Ad Hoc Committee Meeting for "Wearables" Standards on Wednesday, July 16, 1997: Tefralgar A Continental Breakfast (7:30 a.m.) Opening Remarks(8:30 a.m.) Dick Braley Introductions All The IEEE Standards Process J. Isaak, L. Johnson Break (10:00 10:15 am) Terminology.Activities, Outputs Lunch (11:30 - 1:00) Brainstorming "Wearables" Standards All Narrowing the Options Break (2:30 - 2:45) Focus,Support and Responsibilities Action Items/Next Meeting D. Braley Adjourn (5:00 p.m.) As I had expected, it looks as though we will have a good turn out, and we should get some things accomplished with this number of participants. You may register Tuesday evening or early Wednesday morning, at which time you may also pay the registration fee. The meetings will be held on the first floor of the Crown Plaza, and there should be an information desk should you have questions. I look forward to seeing you on Wednesday. Thanks in advance for your participation. Dick Braley (901-375-6535) ------------------------ Dick Braley FedEx 901-375-6535 rcbraley@fedex.com Bob Heile BBN 617-873-4835 bheile@bbn.com Mark Munson GTE 770-804-3960 mmunson@mobilnet.gte.com Steve Case ViA 507-663-1399 scase@flexipc.com Jim Carroll ViA 507-663-1399 jcarroll@flexipc.com Nathan Lemmon Fedex 901-375-6762 nlemmon@fedex.com Dennis Martinez M/A-Com 508-442-4990 martined@macom.com Ian Gifford M/A-Com 508-442-4650 marketing@macom.com Rehmi Post MIT 617-253-9497 rehmi@media.mit.edu Tom Runner McDonnel Douglas 205-922-6713 tia.runner@a90.mort.com Bradley Rhodes MIT 617-253-9601 rhodes@media.mit.edu Roger Hill Xetron 513-881-3415 rogerh@xetron.com Larry Ochs Xetron 513-881-3266 larryo@xetron.com Peter Gallo Symbol Tech 516-738-4989 gallop@symbol.com Joe Dvorak Motorola 954-723-4818 dvorakj@plnt004.comm.mot.com Rod Guy Motorola 847-576-4578 crg005@email.mot.com (Jim Isask IEEE) ------------------- How standards work: PASC Sponsor group: They're the ones that vote on standards (no fee, anyone can get on) When doing actual work, submit a PAR including Scope of project and purpose of work. "Invented" work (never before implemented) is frowned upon. There's lots of logistics -- one big SNAFU is that you need to have a 75% yes-vote -- abstainers will kill you (esp people who became sponsors just to get the proceedings w/out wanting to help). TAG: Technical Advisory Group You can be a study group up till balloting, so getting the PAR early isn't actually necessary. Fastest we could ever do a standard is 9 months, and that's if we knew right off what we wanted. 10/12-10/17 is next meeting time (in Reno) -------------------------- Dick: We want to be international. The only other similar activities known are connectivity / wireless networking. ---------------------------------------------------- Questions: Are we looking at communication with objects? (we should) touch: Use/context: doorknob, gun Intentional: touch-pad, handshake meter: Use/context: display is in area, person with similar taste in area Intentional: Directional (IR) miles: CDPD, GSM, Ricochet, Wavelan Esp need crypto for meters & miles. Also need to make sure you don't interfere with other people's wearables with those ranges. Some definitions: Levels of communication: Wearable Computers -- having one piece of a worn computer talk to another Local area: talk to objects/other computers in a nearby area Nomadicity -- being able to communicate with a base station, sometimes many miles away Humeonics: usability, human interface, total usability Zero distance: capacitance (MIT's PAN, touch) Nano: Smart-card Micro: Meter Macro: Miles Global: Satellite ------------------------- Dick's presentation: Purpose: encourage development of standards for wearables, solicit IEEE support to develop standards Wearable: Computing devices, networking devices, software, and periferals -- worn or carried by individuals -- that will enhance their ability to perform productive work. (Unobtrusive is key to Fedex) We need to include PDA's, smart cell phones, etc. We need to stay at a high level or we'll get bogged down. Wearables markets: Service, medical, maintenance & repair, operations coordination, barcode, military, manufacturing, car rental, inventory, training Manufacturers (partial list): Fedex, symbol tech, norand, ViA Inc., Videx, Rockwell, Telxon, Xetron Standards are important to the end-user so they have options (won't get locked in to a single provider) Potential standards: PAN, Interfaces, system architectures, nomadicity Periferals, power system Are there needs for standards? Concern: can we make a standard that's relevant? Will we be bulldozed by the Microsofts/Sonys out there? Is the field mature enough for us to keep up? (Fedex went out with an RFP for RF communications protocols -- there's nothing available commercially yet that covers their needs) ---------------- Matrix of communications distances vs bandwidth, and see what's available Suggestion: break into groups to identify different areas where we might be able/want to set standards. We want periferals to inter-operate -- that's what we're here for. Do we have to assume that whatever we come up with will be irrelevant within two years? Can we identify things that won't change quickly? Criteria for things to standardize: Will still be valid in two years Things we'll be able to control/affect with a standard ----------- Things to standardize: Yes: PAN (zero-range to micro (~5 meters) -- capacitive, IR, RF) |Physical Interfaces / connectors (computer <-> computer interfaces) |Software interfaces -- Text-speech, dialogue management, GPS, etc More likely to be railroaded through interface No: Periferals No: Nomadicity (5 meters+) -- sounds like it's already being covered elsewhere No: Wearable computers (CPU, operating system, related systems) No: Power/Battery (power-saving in other systems/communications) [Safety/regulation? Would be good to at least perform due diligence with safety groups (SEC29) to mention possible hazards of what we propose] ------------------- PAN (Network whose communication span is 0-5 meters, wireless) IR -- IRDA standard has been done (de-facto from HP) Capacitance -- deferred for further research RF -- requirements: Don't interfere with each other (interference suppression) Components not interfere (dynamic node addition / autonomous node addition) Transmission speed -- auto speed negotiation? Software configuration? (Private eye does 8Mbits/sec -- do we need to worry about such high speeds with this technology?) Authentication / privacy / security I/F protocols Power management 802.11 specification is similar, but higher throughput (1-2Mb) with too high a power requirement. We need to say unequivocally whether 802-11 would work or wouldn't work for our application (breaks our power budget by a power of 10). Our standard has to be feasible for lower powers. (Don't have to set an upper power bound, we just have to make sure it's feasible at higher levels). Should we be under PASC, not under LMSC (Land Mobile Standards Committee, 802)? Also, if we sent this to 802.11, would this isolate the hardware/software interface issue if this stays here. We might want to get together with 802 to get expertise -- we might even want to be under them to get an 802 number, but they may not be interested in just the small subproblem we're interested in. Put the bulk of our cycles into solving the problems we need to solve, but keep communicating with other groups. Could also possibly help write 802, microprocessors, PASC, chapters. In the meantime, keep defining the problem. People we could go to: 802: Good expertise, maybe not quite right focus, well organized PASC: OS, API Microcomputer group: less organized, very open Bus architectures committee: probably not appropriate SAB: sponsor of last resort (superset) -- no structure at all. We should write a project definition: "here's the totality of where we want to get to." Easiest way to do a proposal is to suggest a single approach that solves the problem -- multiple equally good approaches just get in the way of each other. Cost effectiveness is key (fedex needs ~250,000 of these) -- somewhere between $0.05 to $500 Range: We really want a single solution for the 0-5meters (0-10???). You can always fall back on a lower range if you want. PARs should be on the web for many groups (including 802 stuff). Vision statement on interfaces: Steve Case