The initial idea was to make a globe upon which the earth's temperature could be felt by touch linked directly to the Internet via TCP/IP.
n the development process, however, 01 took a slightly different direction. (We would still like to realize the globe idea at some point.)

I will now give a brief description of 01's structure and production process.

[Outline of BeWare01: Satellite]

Some sensorium members are "Weather Tracker" users, a small Mac shareware utility that provides access to a weather server on the www, and the latest temperature data for locales around the world.
This software taught us of various other potential applications of the Internet other than the Web or e-mail. We were grateful for the ideas it provided.

We began researching the state of air temperature and its relation to the Internet, ie. to what extent this information was available on the Net, in mid-June.
For instance, the Weather Underground site, gives the latest temperature data from 5000 locales around the world, but the frequency with which the data was updated was sporadic, and could hardly be called "real-time." The Internet was not being used for weather information universally--so it goes.
(During this survey we learned a tremedous amount from A. Sneider, Tomoaki Sakurai and mailings from the Chikyu Ryutaidenno Club.)

For the time being we concluded that the weather observation network in all countries was not directly linked to the Internet.
Understandably, the guaranteed delay and packet loss of the Internet would be inexcuseable in data for serious uses like airline network control. We learned that there was another network operating besides the Internet for universal international weather information. (cf.1,cf.2)
There is a measuring tool that for TCP/IP Internet attachment already on the market, but until it covers the entire world, the aforementioned concept of the globe linked to real-time temperature data would not be realized.

At that point, we began to check satellite data rather than ground observation networks.

We were surprised by data from stationary satellites such as "Himawari" and "GOES" from the NASA site. Images take by a variety of different cameras is are made public six hours later.
Satellite cameras, much like a peep-hole camera, are not directly linked to the Internet. The labors of NASA staff fill the gap. It is perhaps automated to an extent but the system requires tremendous work and creativity to realize.
This process increased my respect for satellites as well as for the Internet.

Weather data obtainable through NASA's site comes from three sources, the Japanese "Himawari," and the American "GOES-8" and "GOES-9." These do not however cover the entire earth. METEOSAT hovers over Europe, but its only Web data is graphical, which would be difficult to apply so had to be abandoned. Data on the Web page of the Russian satellite centered over the Indian Ocean appears in Russian so likewise had to be abandoned.

Parallel to the data survey, our engineering team conducted experiments in creating an interface upon which changing temperatures could be controlled and physically experienced by touch using Peltier thermoelectric module sensors.

One problem was handling heat exhaust of the Peltier modules. A large portion of the earth when seen from a satellite is covered in clouds. Clouds are cool and the Peltier modules must express that, however characteristic to the modules is that when one surface cools, the opposing surface generates a proportiioal amount of heat. Thus without a sufficient number of radiator plates temperature control would be inffective. Such hardware limitations along with data receiving conditions influenced the final form of the piece in late-July.

After exploring numerous ideas for using stationary satellite data, we decided upon using data from the axial orbital satellite NOAA. NOAA orbits longitudinally making use the earth's rotation to scan the earth's entire surface in approximately one day.
Unlike stationary satellites, these satellites orbit at a relatively low altitude and have a fairly high rate of falls. The present NOAA is the fourteenth to be lauched.

NOAA data is obtainable upon written request on a Web page form and waiting for e-mail notification that the data prep is ready.
The time lag for data sent FTP is irregular, anywhere from one to almost 24 hours. (Japan is painfully slow in releasing and universalizing this kind of public data.)

The BeWare01: Satellite requests data from the NOAA site periodically via the Internet and generates live data in connection with it. Visibile imagery is animated and IR (infrared ray) imagery is changed into temperatur e information and reflected on the top surface of rectangular plate.

The system stops if the Internet is down or if data from the NOAA end stops. (Of course should the request command form change it would also stop. We are praying this does not occur during the course of the exhibition, and have contacted the NOAA site manager on the subject . . . )