- From: Marcos Caceres <w3c@marcosc.com>
- Date: Wed, 22 Feb 2012 13:56:49 +0000
- To: Myriam Leggieri <myriam.leggieri@deri.org>
- Cc: Arthur Barstow <art.barstow@nokia.com>, public-sensorweb@w3.org, Michael Hausenblas <michael.hausenblas@deri.org>, Manfred Hauswirth <manfred.hauswirth@deri.org>
Hi Myriam, On Wednesday, February 22, 2012 at 1:38 PM, Myriam Leggieri wrote: > Hi Marcos, > > On 22/02/2012 13:05, Marcos Caceres wrote: > > we don't even have a simple sensor API or a means to communicate with > > a sensor network… let alone communicating with a kit like an Arduino). > > > a mean to communicate with a sensor network / platform / node is exactly > what is addressed by describing the information involved, using Semantic > Technologies. Such technologies ease the data communication since they > ease the understanding and integration by not relying on ad-hoc schemas > or top-down imposed standards (often not covering all the particular > needs and having low uptake). That's why ontologies, RDF, Linked Data > are important technologies / principles. I agree in principle (and of the principles), but I would not want to shape any decisions based on a particular set of technologies (except for HTML5, DOM4, and WebIDL - which underpin the Web Platform). Other technologies, like RDF and friends, could then make use of the given sensor APIs (but the APIs should be agnostic to the Semantic Web technology or any other technology and just focus on meeting a given set of use cases - namely ones around communicating with something like an Arduino to read pressure sensors, turn on/off LEDs, etc.). > I'm very interested in a discussion about this (as, I guess, many > others), since it would be nice if the Sensor Web and the Semantic > Sensor Web [1] could cooperate and unify their efforts. I agree. It would be good if they could build on each other; and it would be good to get a nice separation of concerns. Right now, I'm personally more interested in just talking to sensors and not so much how those sensors are linked into networks (and nowhere near discussing what possible range of technologies may already be out there to solve the particular problems for a particular audience of developers… which may or may not be addressed by the Semantic Web technologies… point being: we should consider a range of options, because it sounds a bit presumptuous to assume the Semantic Web stuff can address the use cases of a networked set of sensors before we know what the use cases are). Kind regards, Marcos -- Marcos Caceres
Received on Wednesday, 22 February 2012 13:57:25 UTC