Slashdot Mirror
Moving Sensor Data Onto The Internet With SensorML
Roland Piquepaille writes "According to this Sensors article, a new XML encoding scheme may make it possible for you to remotely discover, access, and use real-time data obtained directly from Web-resident sensors, instruments, and imaging devices. By describing sensors using SensorML, anyone can put sensors or sensor data online for others to find and use. And because it's XML-based, it means all this data will easily be searchable. "For example, searching for particular kinds of sensors and data in a particular geographic region, with data collected within a particular time window, will be easy. This has significance for science, environmental monitoring, transportation management, public safety, disaster management, utilities operations, industrial controls, facilities management, and many other activities." In this column, you'll find a summary of the Sensors' story which contains more technical details about the technology. And if you're really interested, please visit the SensorML homepage."
They didn't shorten it to SML because everyone will pronounce it smell. Then half their FAQ will have to explain that smell sensors don't exist yet.
Jason
ProfQuotes
<I>
<have fallen="true">
<can>
<get up="false">
</can>
</I>
It might be difficult to establish trust for these technologies. If I want to know the temperature in some distant city, how can I be sure that the sensor I address is correctly calibrated, or not resting near a air-con outlet? In fact, there would be no way to tell if the damn thing even exists- it could be a textfile sat on the sever or the local tourist office...
Just think of what could be done when Lego updates Mindstorms to use this.
Google on smell sensors faq
Smell with a digital sensor
cyrano sciences
How Smell Sensors Work
. . .
From perusing the SensorML site, SensorML seems geared for satellite-based geosensing and other applications for which significant computing resources are available. I'm interested in small wireless sensor networks, which have very limited computing resources (usually just an 8-bit 8051 or HC08-based MCU, running at 4 to 8 MHz, with about 5k ROM and 1k RAM available). SensorML seems to have a lot of optional fields that could perhaps be eliminated for a "stripped down" version suitable for these types of sensors but, while I'm not an expert in the field, it's always been my understanding that XML=bloat. Can anyone comment on the feasibility of SensorML for small embedded applications?
Seems to me that National Instruments could use this recommendation to develop a web services based version of LabView. It'd be cool to have a loosely coupled, geographically independant sensor network that one could run experiments against. Now if we only had RemoteHandsML.
Pimpin' all the Karma Hoes!
They've invented...SNMP.