Hello Folks, In the following mail, I wrote that I designed a sort of "Soft Emergency Alert System". In this work, my joint researcher, Assoc. Prof. Ogashiwa (Maebashi Kyoai Gakuen College) coded a new XMPP server / client system from scratch. And these server / client codes are kept simple and small, so now we (I and Assoc. Prof. Ogashiwa) have a very good environment to implement new functions as new realtime services by using XMPP protocol suits. And he have leadership to realize "Kyoai Ubiquitous Campus". In this network assisted campus, he built a XMPP testbed using handreds of xmpp enabled iOS devices and his server / client codes, so we can evaluate new realtime services using new functions of XMPP, and deploy the services in real daily campus life. If you OK, we'd like to introduce "these new implementaion of XMPP client / Server system", and "a XMPP testbed in the real life" in the next XMPP summit.
Now we are going to prepare a new research project, called "RSi Project" in Japan. "RSi" stands for "Realtime Service Initiative". In this project we will provide new web site to share Assoc. Prof. Ogashiwa's new codes as open source under the New BSD Licence, and we are welcome to join to reserch / develop / deploy activity with us. From this project, Assoc. Prof. Ogashiwa and Mr. Sato in master course of Keio Univ. and I will join the next XMPP Summit. Best Regards, Yoshihiro Suzuki (hiro.suz...@d3communications.jp) On 2012/01/13, at 15:39, HIro Suzuki wrote: > Hello XMPP Communities, > > I'm very interested in realtime services by using XMPP protocol > suits, and recently I designed a sort of "Caregiving System" for elder > people by using mainly PEP (xep-0163). Now Japan is rapidly becoming > an aging society. So, local governments like city has big issues to > assist or care elder people's daily life. > > In this design process, I have 2 isuues, 1) Event Aggregation, 2) > Event Visualization, if you OK, I'd like to talk about this issues and > my current solution in the next XMPP Summit. > > --- > Abstruct of 2 Issues in Event Signaling > > This "Caregiving System" is a sort of "wide area nurse call system", > or "soft emaergency call system". As a prototype, residential > terminal will be set in handreds of house at where only 1 elder person > lives, and when he or she wants care, he or she push a button on > residential terminal. And this residential terminal has a heat sensor > and a smoke sensor to detect a fire, and a gas sensor to detect a gas > leak. If residential terminal signals some events, events are sent to > call center of the city, and on a terminal in the call center, the > termial displays notification based on events and some usefull > informations (address, map, cantact list, past illnesses...). > > In this design process, I have some opinions about "event > aggregation" and "event visualization" in "PubSub" and "PEP" > specification. > > * Event Aggregation > > In this system, when residentail terminal signals event, signaling > event has only basic informations, as example, terminal id, event > time, event source. (Because simple function is desireble to keep > cheap cost, and informations from residential side is often > insufficent or incorrect, so event gathering/aggregation side needs > some information processingg function. "Collection" of PubSub > specification is now lack of this information processing function. > > * Event Visualization > > I think any events should have essential visual symbol, so when event > is notified, preset visual symbol should be displayed on the terminal. > And when event has some parameters, the client terminal to recieve > notification should have processing function to visualize event > parameters. > > I'm very interested in recent standardization activity of IETF > RTCWeb/W3C WebRTC WGs, above issues, especially "Event Visualization" > may be involved. Now, in RTCWeb/WebRTC WGs documents, they are > focusing symmetric model only (in "Overview" document of RTCWeb: > draft-ietf-rtcweb-overview-02, following "Browser RTC Trapezoid" model > is written, this trapezoid model is symmetric). In order to realize > "Event Visualization", I think it is necessay to discuss asymmetric > model (one side has browser to visualize event, and other side has > only signaling funaction without browser). > > --- > > +-----------+ +-----------+ > | Web | | Web | > | | Signalling | | > | |-------------| | > | Server | path | Server | > | | | | > +-----------+ +-----------+ > / \ > / \ Proprietary over > / \ HTTP/Websockets > / \ > / Proprietary over \ > / HTTP/Websockets \ > / \ > +-----------+ +-----------+ > |JS/HTML/CSS| |JS/HTML/CSS| > +-----------+ +-----------+ > +-----------+ +-----------+ > | | | | > | | | | > | Browser | ------------------------- | Browser | > | | Media path | | > | | | | > +-----------+ +-----------+ > > Figure 2: Browser RTC Trapezoid > --- > > Just for information, I attached one picture of "Overview of 'Caregiver > System'" in this mail. And now, I started to prepare presentation > materials. So it's my pleasure, if you give me a chance to discuss with > you. > > Best Regards, > Yoshihihro Suzuki > (hiro.suz...@d3communications.jp) > > <Overview of Caregiver System.jpg>
