< a proper†y="wo†-core:name" rel = "wo†-core:self"
href="http://somewhere.com/lamps/444">Bob's desk Iamp

Owned by < a †ypeof="foaf: Person"
property = "foaf: name"
href="h††p://somewhere.com/users/222">Bob

</scrip†>
<div rel = "wo†-lifecycle:s†a†echart"
resource = "http://somewhere.com/lamps/444/scxml">
<p>The lamp's state is:
<img src = "swi†ch_on.png" rev="wo†-core:image"
resource = "[_:switchon]" onclick="lamps.switchOff(444)"
style = "cursor:poin†er"/>
< p abou†="h††p://somewhere.com/lamps/444" rel = "wot-
lifecycle:s†a†e"><span †ypeof="wo†-lifecycle:S†a†e" proper†y="wo†-lifecycle:value"
resource = "[_:switchon]">
lamp. switch.on
</span>
</p>
</div>
< div reI= "wot-co re:hasOperation">
To <bxi> <span †ypeof="wo†-core:Func†ionali†y" property ="wot-
core:hasName" resource="[_:blink]">blink</spanx/ix/b> the lamp, you
can click on this bu††on<br/>
<img src="blink.png" rev="wo†-core:api" resource="[_:blink]"
onclick="lamp.blink(444)" style = "cursor:pointer"/>
</div>
</div>
</bodyX/html>
These representations may be provided by the real objects themselves.
This is because some objects may be sold with software means that allow a
representation of their abilities to be provided to external devices. These
representations may be different from the one described above, but different
conversion mechanisms may then be provided.
The representations may also be provided by a third party or by an
administrator. For some types of real objects, representation templates may
also be provided, particularly being downloadable from a website. This is
because, regardless of the manufacturer, a lamp, for example, may have a
very standardized representation. Some parameters may potentially be
modified (dimmer, spectral purity, etc.)
Adaptation devices OWE1 , OWE2, OWE3 may be in charge of
interfacing the real objects Objl , Obj2, Obj3 respectively, with the
communication network N. These adaptation devices may have the function of
making the representations associated with each object available.The adaptation devices may form a 1- 1 association with the areas, as
depicted in Figure . They may also be embedded into the objects themselves;
in such a case, an adaptation device is associated with a real object.
Other implementations may, however, also be conceived.
Thus, in Figure , the adaptation device only interfaces with a single real
object each time, but it is conceivable to have a single adaptation device for
multiple real objects. This is obviously the case if a 1- 1 association is
maintained between area and adaptation device and if an area includes
multiple real objects (which it generally does); however, it is also possible to
have multiple adaptation devices within a single area and for one or more of
them to interface with multiple real objects, for example based on their type,
nature, or manufacturer.
The representations may be saved within a storage means. These storage
means may be colocated with an adaptation device. In the event that the
adaptation device is integrated into the real object, its representation may also
be embedded into the memory incorporated into the object.
It is also possible to associate with the adaptation devices managing
multiple real objects, a database D3 to enable it to maintain the references
(and addresses) of these handled objects.
Furthermore, a centralized database D2 may be used to form a directory
of the real objects. For example, whenever an adaptation device connects to
the communication network, it may update that directory D2 by inserting
within it the references (and potentially the addresses, characteristics, etc.) of
the real objects that it manages.
The terminal control device TC is capable of searching for all or some of
the available real objects. To do so, it may query the database D2. It may also
query the adaptation devices OWE1 , OWE2, OWE3.The way in which the terminal control device becomes aware of the
adaptation devices OWE1 , OWE2, OWE3 is outside the scope of the
invention. Just a few possible implementations may be mentioned here: the
adaptation devices may regularly send over the communication network N
messages reporting their existence and containing their address. Alternatively,
they may update a registry collecting the real objects available within one area
or a given larger-sized area, and their addresses.
It may also be possible for a given object to inform the terminal control
device TC of the other objects of which it is aware. This implementation has
the advantage of adapting to "friendly objects": for example, an amplifier will
be aware of the existence (and technical characteristics) of the DVD player,
speakers, etc. to which it is normally connected.
The real objects determined in this way by the terminal control device TC
may be presented on the human-machine interface of the terminal TC. (?)
In Figure 2, the objects are presented in a graphical form within the
object area ZO of the touchscreen E. From top to bottom, they are: a
telephone, a medium-sized screen, a computer, a larger-sized screen
(television set), a digital photo frame, and a radio.
The set of real objects may always be offered to the user U, regardless of
the application selected.
However, it may also be possible to offer a filter.
This filter may be based on the type of application and on the points of
interface associated with that application. For example, for a video output
point, only the real objects capable of offering a video display may be
proposed.
This filter may also be based on the geographic zone.This filtering mechanism may be similar to the one that is used for
automatically matching the services' points of interface and the available
operations of the real objects.
The invention has processing means SM for performing at least one
match between a point of interface associated with the selected service and at
least one available operation contained within the representation of a real
object and associated with a current state of that real object.
In more concrete terms, in the example in Figure 2, this involves locating
from among the real objects depicted in the left-hand area ZO, the objects'
operations available as a function of their current states, which correspond to
the three output points "video", "photo", "RSS" for the service "Facebook".
This matching may be performed manually.
The user may click o n one of the objects depicted in the object area ZO
and drag it into the central area ZM across from the desired output point. He
o r she may, for example, drag the digital photo frame this way onto the
"photo" or "video" output point.
The processing means may then provide any verification function, in
order to check that the object selected by the user does have a n operation
corresponding to the interface point.
For example, it may involve verifying that the digital photo frame does
have the technical characteristics to play and display videos, if it is matched
with the "video" output point of the application "Facebook".
Additionally, it may be possible to make a n operation available for a
selected real object, but in a different state from the current state. For
example, the television is selected to display videos from the application
"Facebook", but it is off. The processing means SM may then detect this
situation based on the representation associated with the "television" object,and offered †o cha nge its state in order to make the "play video" operation
availa ble. The processing means SM may determine the actions to perform on
the real object in order to put it into that state; or in the "nea rest" state if the
object has more than one state in which the operation in question is available
(the "nearest" state may be the one that req uires the least action, or the least
costly actions) .
The processing mea ns SM may also perform a match automatical ly.
This matching may be performed by pairing off operations and points
interface that have a strong sema ntic correlation. To do so, semantic values
may be assig ned to the operations and points of interface, within thei r
representations.
The real objects' representations may contai n a keyword associated with
each operation . This keyword may be the identifier of the operation itself, or
may complement that identifier.
For the lamp, this keyword may be "blin k" and be inserted into a
"resou rce= " line.
Keywords may also be associated with the points of interface of the
available services. These keywords may then be saved with the services'
representations.
In the example of the service described above, this keyword is inserted
into an attribute "kind". It takes two val ues in this example, "light" and "text".
This keyword may be the identifier of the point of interface itself, or may
complement that identifier.
One simple implementation may consist of matching operations and
points of interface that have the same keyword .However, other mecha nisms are possi ble in order to seek to optimize the
sema ntic correlation between operations and point of interface.
It is also important to take various factors into accou nt:
one semantic val ue may be more general tha n the other. For example,
an operation may be "display image", while the service has an output point
"display photo". It may be beneficia l for the processing means SM to be able
to determine that an image is a generalization of the concept "photo", and
that the match is possible even if the keywords are different.
Keywords may be synonyms. For example, "photo" and "photograph",
"video" and "movie". This situation may particularly occur if equi pment and
services from different man ufacturers are used, each one usi ng its own
voca bulary. In such cases, it may be useful to be able to bypass purely
syntactica l matching based on identical keywords.
A need may therefore exist to perform a finer correlation . To do so, the
processing means SM may use an ontology. This ontology may be avai lable
on the web, or may be more loca l and delivered with the processing means
SM. In the latter situation, updates may be provided in order to incorporate
cha nges, new manufactu rers, new services, new equipment, etc.
This ontology may be described in different ways, based on different
models, particula rly RDF schema, OWL that come from the work of the W3C.
By using an ontology, it becomes possible to make correlations between
synonymous concepts ("video", "movie", etc.) or close ones ("image", "photo",
etc.) .
For example, the point of interface "scintillate" that is fou nd in the
examples given above may be matched with the operation "blink" from the
lamp example.
The automatical ly offered matching may potential ly be modified
afterwa rd by the user U. It is displayed in the area ZM of the termina l T, andthe user U may, by the human-machine interface, select the match in order to
delete, edit, or configure it.
For example, he or she may configure the resolution with which the video
is to be displayed by the television screen. His or her choice may be distinct
from the automatic choice.
Sometimes, the same point of interface may be matched with multiple
operations.
If so, multiple options may be taken by the processing means SM:
- all of the possible choices may be offered to the user via the human-
machine interface of the terminal T. He or she may then select his or her
choice similarly to what was described for manual matching, though with the
object area ZO offering only the objects that have been automatically
matched. For example, for the "video" output point of the Facebook
application, the object area ZO will only show objects that can display a
video: television set, computer monitor, etc. in the appropriate state ("on").
- it may be provided to use a profile of the user U. This profile may
comprise preferences, particularly with respect to his or her preferred
equipment. It may thereby be possible to indicate that the user U prefers to
watch his or her videos on television. Otherwise (television off), he or she
prefers to see them on the computer monitor. This profile may be entirely set
by the user, but it may also be provided that the log of the user's choices and
actions will be used to at least partially define it.
It may additionally be proposed to match a point of interface with an
operation of an object that is not available in that object's current state.
This proposal may make it possible to enhance the possible choices for
the user, or to propose at least one choice whenever none would otherwise be
possible.For example, if no monitor is turned on when a videophone call is made,
it may be proposed to turn one on. Additionally, in order to view a video from
the application "Facebook", if only the digital photo frame's screen is on, it
may be proposed to turn on a larger-sized screen such as the television set.
Once again, all of the possible choices may be offered, or just a more
limited sample.
Delimitation of the choices may be carried out as a function of a profile
of the user.
It may also be carried out based on a notion of cost.
This cost may take into account
-the number of actions needed to put the real object in question into a
state in which the operation is available;
- The duration of these actions (for example, turning on a computer to
display the video from an incoming videophone call might not be suitable,
given the necessary boot time that this would involve)
-the energy consumption incurred by each action, etc.
The table below depicts different situations that may arise for a telephony
application for people with hearing difficulties, as previously described.
The right-hand column indicates 3 situations, and the left-hand column
indicates matches for these situations. Only the 3 output points "Ringing",
"Blink" and "Display" mentioned above are taken into account, and only the
match's real object and not the operation's is indicated, as the operation is
obvious to deduce from it.
Normal situation Ringing &#8594; telephone
Blink &#8594; lamp
Display &#8594; TV
TV off Ringing &#8594; telephoneBlink &#8594; lamp
Display &#8594; digital photo frame
TV off, telephone off-hook Ringing &#8594; alarm clock
Blink &#8594; lamp
Display &#8594; digital photo frame
The processing means SM may check that all of the mandatory points of
interface are indeed matched with a real object's operation. If not, it may
notify the user U that he o r she must complete its configuration if the matching
is manual, o r that the service cannot be deployed.
Finally, the inventive device comprises activation means LM for deploying
the matches made in this way, by connecting the service and the real object in
question.
To do so, the activation means may communicate with the real objects
directly or by means of the adaptation devices OWE1 , OWE2, OWE3 by using
the protocols, languages, and mechanisms that are available.
In particular, the mechanisms DPWS (for "Device Profile for Web
Services"), REST (for "Representational State Transfer"), WDSL 2.0, etc. may be
used.
This DPWS mechanism is particularly described in the links
http://schemas.xmlsoap.orq/ws/2006/02/devprof/ and
http://download. microsoft. com/down load/b/5/3/b53ea430-dbe5-440c-
a3Q8-df97bl 0280b7/introducinq dpws.pdf
Using these technologies, the activation means LM may establish the
communication between the services' points of interface and the real objects'
operations.To do so, the application may use the API provided by the adaptation
device OWE1 , OWE2, OWE3, described in RDFa in the HTML representation
that it itself provides.
Furthermore, the activation means LM may control, in real time, the
availability of the objects and the state of the connections between services
and real objects If the connection is interrupted, they may thereby send a
message to the processing means SM so that they can trigger a new match
with another real object that is available.
The activation means may maintain the relationship with the real object
via the adaptation devices OWE1 , OWE2, OWE3.
These devices may, for example, have a subscription mechanism to
which the activation means LM may subscribe in order to be automatically
informed of the real object's events. Alternatively, the activation means may
regularly send requests in order to be assured that the real objects are still
available.Claims
1) A device for deploying a service onto an environment's set of real
objects (Objl , Obj2, Obj3), comprising means of interaction (T, TC) with a
user (U) in order to select said service from among a set of available services,
each available service being associated with points of interface,
means of storing representations (OWE1 , OWE2, OWE3), each
representation being associated with a given object from among
said set of real objects and exhibiting associations between available
operations and possible states for said given object,
processing means (SM) for performing at least one match
between a point of interface associated with said service and at least
one available operation contained within the representation of a real
object and associated with a current state of said real object.
and activation means (LM) for deploying the at least one match
by connecting said service and said real object.
2) A device according to the preceding claim, wherein said points of
reference belong to a group comprising input points, output points, and event
points.
3) A device according to one of the preceding claims, wherein said
points of interface may be either mandatory or optional, and wherein said
processing means (SM) perform the match for each of said service's
mandatory points of interface.
4) A device according to one of the preceding claims, wherein said
means of interaction (T, TC) comprise a communication terminal (T) equipped
with a display means operative to present to said user at least one part of saidset of available services, and an input means for selecting said service from
among said at least one part.
5 ) A device according to the preceding claim, wherein said display
means and said input means are a touchscreen.
6 ) A device according to one of the preceding claims, wherein said
match is performed by pairing off operations and points interface having the
same keyword.
7 ) A device according to one of the claims 1 to 5, wherein each match is
performed by pairing off operations and points of interface having a strong
semantic correlation.
8 ) A device according to one of the preceding claims, wherein said
means of interaction are additionally operative to present said user with at
least some of said performed matches and to enable him or her to select one
or more matches.
9 ) A device according to one of the preceding claims wherein whenever
multiple matches are performed for a single point of interface, the
determination of which match is to be activated is performed as a function of
a profile of said user.
10 ) A device according to one of the claims 1 to 8, wherein whenever
multiple matches are performed for a single point of interface, the
determination of which match is to be activated is performed as a function of
each of said multiple matches.