Opening Tatsuya Hagino Nobuo Saito -- NTT DoCoMo A brief history of wireless communication system IMT-2000 Q: Can mobile phone itself have a IP address, and is accessed from Internet ? Yes. NTT already provide some products that have IP-address, and future products will be the same. It will supports both of fixed address and dynamic one. Q: For asymetric transfer protocol, spec is 384k bit/sec for down, How about Uplink ? It will 384k bit/src for both up and down link, but it can be fall-back to more slower speed that the service needs. -- TOYOTA MONET Standerization of Map information How to exchange information about location in WWW? Q: How is Map information updated? (Ex. New gas station, New store... etc.) Is there any idea to do it automatically? It depends on contents provider, or author of the page. Q: I think that position of gas station is got by data of latitude and longitude, but on the web site, similer data is not exists. At the moment, there is no standarization for map information, longitude, and so on. We need some standard. Q: Are you try to suggest a new tag for position in HTML? I want to discuss what kind of way is best way for position data. HTML TAG? MIME-format? or some other format? Q: How many people useing MONET? About 1000 subscriber Q: What kind of protocol is used over PDC ? It's a original very simple protocol, similers to PTY or BSC. Q: Why don't you use DICS instead of PDC? PDC is for on-demand information. DICS is just given to receiving only traffic information. Q: How is MONET system know when new e-mail is arrived? When user select some kind of button, MONET EUC trys to get new e-mail information. Q: Do you wish to define new HTTP method to put the location infrmation, No, One idea is new HTML tag to display a location. I think it's very convinent if uniform definition of location is fixed. Q: How can we send e-mail with position data? Is there a special application? Then, how to get the application, is it free? It's available only MONET sevice subscriber. They can operate the e-mail via MONET member only site. -- Panasonic Motion Pictures on Web Mobile Visual Phone Q: mobile-gateway sends request to the client. Which is this? WWW server request broadcasting to the mobile-gateway. In this case mobileq-gateway is a HTTP server. When the requiest is successed, get back "OK" responce. Q: The mobile-gateway make initial connection to the client, How do you do that? It's a uni-directional broadcasting method. So, there is no acknowledgement. Q: To communicate with mobile client, dosen't server use HTTP? No, it's a uni-directional protocol Q: Are you already defined a protocol? What's name? Yes, There is no name, now. Q: Do you wish to push this protocol as standard? Yes, I hope so. Q: Do you know about idea proposed by Marinba? DRP (Duplication and Replication Protocol) proposed by Marinba. I think it's a kind of phigical compression method. We propose a logical compression method. A merit of using logical complession is, I think server is much more lighter than Marinba proposed one. When using DRP, the server must have entire HTML version. So, this management seems to be very heavy. The second issue is that DRP cannot apply broadcasting mechanism. Because, DRP have to request the serve client's own HTML version. So, the server generate and get back differencial information. -- Microsoft Corporation Using XML, HTML, and Scripting for Offline and Interactive Mobile Access T24 3:15 Q: Do content providers need to migratiuon from HTML to XML? Yes, XML and XML-Data provides solution for desktop computing. It's usefull for mobile. Q. Do you plan some way of registration of client? Not yet. Q. XML-Data ... -- Lotus Development Some Lessons Learned from Mobile Access ... -- WAP Forum WAP desires collaboration with W3C T04 0:30 Q: Why can't you use HTML HTML is not proper for obile device. There are problems that HTML can not solve. Q: How can we use WAP related technorgies? licence? Join WAP Forum and access all specifications. Q: Is it possible conversion from WML to HTML automatically? By using proper stylesheets, it's possible. ... Q: Is there any difference between pure WML and one generated by proxy? Both of them use same DTD, and have different quolity. Q: To make new standards is huge process. Implovemrent of devices will solve the problems. differences of device charactaristics is large... -- Unwired Planet Technical issues of the wireless web ... Q: Contents are important not only for wireless. Dynamic information. I Agree. .... Q: If existing web standards are hard to support, use e-mail and ascii display. It's not enough. -- Nokia Mobile Phone T19 0:20 Q: WAP technorgies and licence WAP spec is open, but there is IPR to use the spec. Q: Dose WML have telephony function It is application. WML-Script can access telephony API. Q: How do you think about Java? .... Q: How about Voice Browser? That is one solution, but it's not applicateable all clients. Q: I'm afraid that WAP don't care continuas change of web. No Q: How about Java? Java is different. Q: WML dose not conform to existing W3C standards, i.e. logical structre and styles are not separated. We have taken best works fron exsisting specifications. We really want feedbacks for next version. Q: If I want to change WAP specifications completely... We have WAP Forum. It is opened for all companies. .... -- Dave Raggett Voice Browser Q: Where do voice recognition It depends on device. if you are using phone, it will be done on proxy server. If you are using desktop, it will be done on the PC. Q. How do you think about summalization It will be done by style sheets and scripting. ----- Acccess T05 4:00 Q: Compact HTML is just a subset of current HTML, but it includes script for Audio/Visual, then it loses advantage. That point shuld be discussed from now. current specufucation is just a subset of HTML 4.0. We should be carefull to add new feautures. ... There might be a varaety of mobile devices, one important thing is to make a negotiation standard by using user-agent attribute - screen size, which function is abailable, memory capaity... something like that. Q: Is it specification CHTML? No, Compact HTML is a markup language. But still I beleve Compact HTML target area is very high volume, so I want to make a standard in this area. Q: Colors, Style sheets, ... Small device have a small screen spaces, but network speed is becoming very fast. In this situetion, to include such new feauture is very difficult. Q: Have you tryed programatically fall-back from HTML to Compact HTML. not yet. Q: To read contents such as e-mail is difficult. It is related to how to make application based on Compact HTML. But typically, application can be made based on Compact HTML. Q: history based navigation It's important to make application UI design carefully. Q: If programatically conversion fron HTML to CHTML is possible, Why don't you want to do that by proxy? .... -- NTT DoCoMo New Interactive Media over the Mobile Communication Network Q: How the content provider generate pages? This is just Compact HTML. Content provider will write text based primitive page. Q: Is there any SDK or tools to convert existing pages? No, we provide a guidline. Q: What kind of services are proper? It's educational stage. Q: To support various clients is hard. Nobody can say that thisi is the best way, The market will decide. Q: What happen when voice call? This is telephone, not web brwser. Telephone call gets priality. You can re-connect after phone call. Q: It looks like AT&T pocket... AT&T pocket is oriented business users, NTT is oriented general use. Q: How it receive push contents? The structure of this service is very simmiler to messanging service. -- Project Xanadu (Prof. Ted Nelson) A Compact Reference Structure for Mobile and Ubiquitous Computing .... -- University of Helsinki (Prof. Martti Tienari) Mobile Computing Based on GSM: The Mowgli Approach Q: wired network is big. wireless netwoek is small. Consider 10 years, wireless will be big, too. Q: wireless packet less is because of noise. .... Q: use HTTP/1.1, it will 5 times speed up. Mowgli can 2-20 times speed up. It depends on data. Q: CDPD & GSM Not be tested yet, but Mowgli will be usefull. Q: TCP/IP for wireless is available. Isn't it usefull? .... Q: relation of WAP This is research. laptop v.s. mobile phone -- Productivity Works Accessibility - A Business and Market Imperative .... -- W3C (Henrik Nielsen) - HTTP/1.1 performance gains over HTTP/1.0 - HTTP-NG - introduction and overview .... Q: HTTP is not needed on wireless current HTTP may be not best, but why do you need new protocol? Q: Don't you trust gateways? religious .... ------- Chris Lilley CSS @media feautre .... -- Free discussion Proposed agenda: (1) HTML v.s. XML HTML extension Voice Browser Compact HTML CSS WML XML-Data (2) proxy (gateway) v.s. direct (3) End User v.s. System Manufacturer (4) specialized v.s. general (4) Fixed Spec v.s. Extensible Spec (5) Wireless Protocol v.s. TCP/IP (6) Working Group v.s. Interrest Group (7) W3C New Domain v.s. User Interface Domain -- Criteria for Success Addressing the concerns of members with respect to mobile access Consensus about the importance and necessity of mobile environments Requirements for (next versions of) HTML, CSS, HTTP, and/or other standards/recommendations Preparation for WG and/or IG proposals if necessary 1. What are people doing for wireless? What people are intending to do with wireless? users What people have done? not yet We have not yet given yet. We need feedback. devices, services and connections Usage group should look at wireless as well. 2. Negotiation between server and client How you describe send capability of devices to servers Current status: HTTP content negotiation --> fuzzy CSS media type definition needs to be more detailed. How do we register new devices? URI solution --> use METADATA clients --METADATA--> servers ??? P3P may be used. We might be able to use METADATA. 3. How do you graceful degradation? initial content ---degrade----> mobile CSS, scripting proxy The way of degradation should be transparent to authors. Author "should" specify the way of degradation. Author shoule be able to disable certain degradation. Author should give up to think how content looks. Clients should tell servers the limitations. "degradation" --> "transformation" Color problem. 4. HTML or XML. HTML, Compact HTML, WML, .... Matter of transition. Political issue. server content: HTML or XML proxy -> client: XML? Wireless devices cannot decide HTML or XML. 5. Do we need separate content on the server? We prefer to have only one content. Who prefer? One URL What is content? Documents and graphics may be different. We need flexibility. We should not restrict to have more than one content. "SAME content" 6. Wireless requirement --> Internet world requirement Broadcasting 7. Ordinary phone is also a mobile device. Coffee maker, car,... Should we cover them here? General architecture 8. proxy or not purpose for proxy Use style sheet. Proxy is useful for many purposes. XML + XSL --> proxy --> mobile device document + script --> proxy --> yes if needed It should be possible to do by proxy. server can do? NEXT STEP *. Combine HTML, HTTP-NG, CSS3, XSL, XML, RDF, .... to sove mobile problems. We do not need to invent new things. Some are underway. *. Do we need working group? Like internationalization working group. Working group does not need to produce specifications. Generating a lot of working groups is difficult. Produce guideline. A strong interrest group for going through existing specifications. By 24th April, send list of specifications and topics that the group wants to pursue, and a list of individuals at each company (names & contact information) that are interested in reviewing the above. www-mobile@w3.org