METHOD AND APPARATUS FOR A PORTABLE INFORMATION AGENT
A method and system for providing enhanced intelligent agents that are moveable by a user from a browser application to a desktop. In specific embodiments, agents provide advanced interactive graphics and communications back to a server.
This application claims benefit of priority from provisional patent application 60/202,736 filed May 8, 2000, incorporated herein by reference.
This application claims benefit of priority from provisional patent application 60/202,731 filed May 8, 2000, incorporated herein by reference.
This application claims benefit of priority from provisional patent application 60/202,730 filed May 8, 2000, incorporated herein by reference.
This application claims benefit of priority from provisional patent application 60/225,956 filed Aug. 17, 2000, incorporated herein by reference.
This application claims benefit of priority from provisional patent application 60/230,341 filed Sep. 6, 2000, incorporated herein by reference.
This application is a continuation application of U.S. Ser. No. 09/852,971, filed May 8, 2001 and entitled “Method and Apparatus for Portable Information Agent,” now U.S. Pat. No. 7,577,700, issued Aug. 18, 2009, and is incorporated herein by reference. This application is also related to the following commonly assigned patent and co-pending applications: U.S. Pat. No. 7,472,157, issued Dec. 30, 2008; Ser. No. 09/852,979, filed May 8, 2001 (abandoned); Ser. No. 09/954,724, filed Sep. 12, 2001; and Ser. No. 10/381,524, filed Aug. 25, 2003.
COPYRIGHT NOTICEPursuant to 37 C.F.R. 1.71(e), applicant notes that a portion of this disclosure contains material that is subject to copyright protection (such as, but not limited to, source code listings, screen shots, user interfaces, or user instructions, or any other aspects of this submission for which copyright protection is or may be available in any jurisdiction). The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or patent disclosure, as it appears in the Patent and Trademark Office patent file or records, but otherwise reserves all copyright rights whatsoever.
FIELD OF THE INVENTIONThe present invention relates to the field of information and/or data handling methods and systems. In specific embodiments, the present invention involves methods and/or systems directed to providing distributable active content that can exist in a variety of software environments.
BACKGROUND OF THE INVENTIONFamiliarity with information and data handling methods and techniques is characteristic of practitioners in the art and is presumed of the reader. At the present time, many people are familiar with accessing information over a data network. The WWW is a public data network that is becoming increasingly used for accessing multi-media information. This information can be one-way, passively experienced information, or two-way information including two-way text, audio, or video data.
At the present time, there is a desire to enrich the user experience. One particular aspect of typical WWW interactions is that interactions take place within the confines of an application, such as a browser. In order to access the information, a user must be at a computer system with a particular type of application for user access. Generally, an interactive application is limited to a particular platform, such as a particular operating system or information handling device.
The Internet comprises computers and computer networks that are interconnected through communication links. The interconnected computers exchange information using various services, such as electronic mail, ftp, the World Wide Web (“WWW”), and other services including secure services. The WWW service can be understood as allowing a server computer system (e.g., a Web server or a Web site) to send Web pages of information to a remote client computer system. The remote client computer system can then display the Web pages. Generally, each resource (e.g., computer or Web page) of the WWW is uniquely identifiable by a Uniform Resource Locator (“URL”). To view a specific Web page, a client computer system specifies the URL for that Web page in a request. The request is forwarded to the Web server that supports that Web page. When that Web server receives the request, it sends that Web page to the client computer system. When the client computer system receives that Web page, it typically displays the Web page using a browser. A browser is a special-purpose application program that effects the requesting of Web pages and the displaying of Web pages.
Currently, Web pages are typically defined using a Hyper Text Markup Language (“HTML”) or similar language. HTML provides a standard set of tags that define how a Web page is to be displayed. When a user indicates to the browser to display a Web page, the browser sends a request to the server computer system to transfer to the client computer system an HTML document that defines the Web page. When the requested HTML document is received by the client computer system, the browser displays the Web page as defined by the HTML document. The HTML document contains various tags that control the displaying of text, graphics, controls, and other features. The HTML document may contain URLs of other Web pages available on that server computer system or other server computer systems.
The World Wide Web is especially conducive to providing information services over the internet. Services can include items (e.g., music or stock quotes) that are delivered electronically to a purchaser over the Internet. Services can also include handling orders for items (e.g., groceries, books, or chemical or biologic compounds, etc.) that may be delivered through conventional distribution channels (e.g., a common carrier). Services may also include handling orders for items, such as airline or theater reservations, that a purchaser accesses at a later time. A server computer system may provide an electronic version of an interface that lists items or services that are available.
SUMMARYThe present invention, in various aspects, involves a method and/or system and/or apparatus for providing an enhanced user interaction in an information processing environment. In various specific embodiments, aspects of the invention enable the presentation to users of a portable information agent (PIA) (at times, referred to herein as an Envoii™) which can provide enhanced user interactions with graphical objects. Particular aspects of the invention further comprise systems, components, and/or methods allowing an agent to be portable over different platforms.
In the present discussion, information available over a public network may be referred to as contained in documents or presentations or compositions. It should be understood that the terms information or document refer to any type of digitally-encoded data that can be presented or transmitted by a computer or other digital device including, but not limited to, text, graphics, photos, executable files, data tables, audio, video, three dimensional data, or multimedia data that is a combination of any of these.
In a further embodiment, the invention comprises a new method for allowing an agent supplier to enhance and track user interaction with an agent and for communicating information between an agent supplier and an agent. In a further embodiment, the invention comprises a new method for allowing an agent to be moved from a browser application to a desktop or to another platform. In a further embodiment, the invention comprises a new method for tracking and reporting back user interactions with an enhanced agent. In a further embodiment, the invention involves a new method and system allowing an agent to move from an initial application to a new location without requiring specific user input. In a further embodiment, the invention involves a new method and system allowing composeability of PIA (or Envoii) objects, by allowing a one Envoii agent to be connected to another Envoii agent, thereby providing additional functions.
In further embodiments, the present invention may be understood in the context of user systems in communication with external data systems over a communication media. An important application for the present invention, and an independent embodiment, is in the field of providing a persistent object that can be initially accessed through a browser and that can move to other software platforms, such as other programs, a desktop, or other devices. In particular embodiments, services according to specific embodiments of the invention can be accessed using an agent over the Internet, optionally using Internet media protocols and formats, such as HTTP, RTTP, XML, HTML, dHTML, VRML, as well as image, audio, or video formats etc. However, using the teachings provided herein, it will be understood by those of skill in the art that the methods and apparatus of the present invention could be advantageously used in other related situations where it is desirable to have a persistent agent.
The invention and various specific aspects and embodiments will be better understood with reference to the following drawings and detailed descriptions. In different figures, similarly numbered items are intended to represent similar functions within the scope of the teachings provided herein. In some of the drawings and detailed descriptions below, the present invention is described in terms of the important independent embodiment of a system operating on a digital data network. This should not be taken to limit the invention, which, using the teachings provided herein, can be applied to other situations, such as cable television networks, wireless networks, etc. For purposes of clarity, this discussion refers to devices, methods, and concepts in terms of specific examples. However, the invention and aspects thereof may have applications to a variety of types of devices and systems. It is therefore intended that the invention not be limited except as provided in the attached claims. Furthermore, it is well known in the art of internet applications and software systems that particular file formats, languages, and underlying methods of operation may vary. The disclosure of a particular implementation language or format of an element should not be taken to limit the invention to that particular implementation unless so provided in the attached claims. The invention will be better understood with reference to the following drawings and detailed description.
Furthermore, it is well known in the art that logic systems and methods such as described herein can include a variety of different components and different functions in a modular fashion. Different embodiments of the invention can include different mixtures of elements and functions and may group various functions as parts of various elements. For purposes of clarity, the invention is described in terms of systems that include many different innovative components and innovative combinations of innovative components and known components. No inference should be taken to limit the invention to combinations containing all of the innovative components listed in any illustrative or example embodiment in this specification. The functional aspects of the invention that are implemented on a computer, as will be understood from the teachings herein, may be implemented or accomplished using any appropriate implementation environment or programming language, such as C, C++, Pascal, Java, Java-script, HTML. XML, dHTML, assembly or machine code programming, etc. Source code examples used herein are given for example purposes. It will be understood to those of ordinary skill in the art that many different source code examples could be used to implement aspects of the invention. All references, publications, patents, and patent applications cited herein are hereby incorporated by reference in their entirety for all purposes.
According to specific embodiments, the present invention extends and transforms user interactions over the internet by providing portable information agents (PIAS) that can be initially accessed in a first interface or window (such as an HTML browser window or an email message window), but that can move or be moved to another software location, such as a desktop. Once moved to another location, a PIA according to specific embodiments of the present invention, maintains for the most part the functionality and connections that it had in the initial location.
1. User Interaction Methods and OperationAccording to specific embodiments, the present invention can be understood with regards to a user's experience interacting with a PIA. This discussion will therefore begin with the user interaction, which contains many novel elements. Following this discussion is given particular examples of implementation details according to specific embodiments of the invention. Many of these details will not be visible to the user and it will be understood from the teachings herein that other implementations are possible within the scope of the invention.
As will be further understood from the discussion below, the technology according to specific embodiments of the invention allows active content to be brought to the desktop without application installations. Envoiis can be dragged from the browser to the desktop or to any Envoii-enabled place. Envoiis can pull data from external sources in response to a user's actions or can be designed to move to a new location without requiring user action.
The present invention has thus far been described in terms of general methods and devices. The previous description is a full and complete description sufficient to allow an ordinary practitioner in the on to make and use the invention as described. It will be understood to those of skill in the art from the teachings provided herein that the described invention can be implemented in a wide variety of specific programming environments and logical systems (such as UNIX, Windows, Solaris, Oracle, etc.) using a wide variety of programming languages such as SQL, Visual Basic, Pascal, C++, Basic, Java, etc.) and wide variety of file formats.
What follows are descriptions of example systems and methods that embody various aspects of the present invention and that describe further novel aspects of particular embodiments. The following discussion is included, in part, in order to disclose particularly preferred modes presently contemplated for practicing the invention. It is intended, however, that the previous discussion and the claims not be limited by the examples that follow. It is further intended that the attached claims be read broadly in light of the teachings provided herein. Where specific examples are described in detail, no inference should be drawn to exclude other known examples or examples described briefly from the broad description of the invention or the language of the claims. It is therefore intended that the invention not be limited except as provided in the attached claims and equivalents thereof.
2. Architecture Enabling Envoiis According to Specific Embodiments of the InventionIn further embodiments, the invention comprises a distributed component architecture that facilitates methods discussed above. From the teachings provided herein, it will be understood that this is one example architecture and that other architectures according to specific embodiments of the invention as previously described, are possible. This architecture can be understood as based on an aggregation model as opposed to an inheritance model. The design offers maximum flexibility and adaptability to environments that cannot always be predicted, without placing undue constraints on what is possible.
In
According to specific embodiments of the invention, the architecture can be understood as a node-based architecture, with a number of “Envoiis.” Envoiis, as described further herein, include active components that can be either MetaVoiis, PlaceVoiis, or Envoiis. Envoiis can be of various types, generally with various functionality and with particular and characteristic visual components. For example, a BusinessCardVoy is illustrated in
An example implementation of an interactive graphical object or agent (herein referred to as an Envoii) according to the invention provides the following services: (1) Envoiis, using built-in traversal capabilities, can perform service retrieval. This means one can ask an Envoii: “Find me service Foo” and the Envoii will return a reference to a service of type Foo, if one is available. In particular embodiments, service requests are propagated from child Envoiis up to parent Envoiis. In some embodiments, atop level parent will be an external server that provides a library or references to all available services. (2) Effectively, the semantics (real-world behaviors like rendering, animation, etc.) are implemented through Envoii services. (3) Envoiis provide support for an arbitrary number of connection points, referred to herein as “ports.” These ports can be named, and provide type signatures, etc., in the form of interface specifications. (4) Envoiis are constructed and added to a tree composition. (5) Envoiis are connected to one-another through the use of objects called “connections.” Connections are generally from port to port, and have explicit references to two ports. (6) Mechanisms are provided for search and traversal of Envoii hierarchy.
Generally, relations between Envoiis are based on the same connection mechanism. Interfaces—Each Envoii supports a set of interfaces identified with a unique interface ID. All Envoiis will inherit default Interfaces and add their own specific Interfaces. When a client Envoii wants to establish a connection to another Envoii's Interfaces, it needs to have a reference to that server Envoii. Either it already has that reference (for example because it structurally knows the server Envoii) or it will get the reference through the Service Discovery mechanism. Using that reference, the client Envoii will query the server Envoii to check that the Envoii actually supports the desired interface. At that point the communication between the Envoiis can be established. The communication is asymmetric in the sense that it has a client/server aspect. Symmetric communication can be established between two Envoiis by establishing two connections. The standard Interfaces include the Service Discovery support, symmetric communication protocols, etc.
Ports—Both ends of the communication are managed symmetrically by ports using a channel. Because an Envoii can have several connections to the same interface, each connection needs to have its own port. In a sense, a port on the server side is an instantiation of the interface. The port on the client side it conceptually similar to a proxy.
Channel—The actual communication between the Envoiis will be encapsulated by the ports to travel through a communication channel. The channel is responsible for transporting the interface method calls and their parameters. It also transports direct communication between the ports themselves. The type of channel transport will depend on the relative location of the Envoiis (in process, inter process, remote). Envoii (or Enode) references are unique inside of their identification space.
Place Voiis
In particular embodiments, an architecture according to the invention further comprises PlaceVoiis and MetaVoiis as described below. PlaceVoiis can be understood as a type of Envoii that allows other Envoiis to operate in a particular place. A computer system that first encounters a graphical Envoii through a particular browser (such as Netscape Navigator™, for example), will receive a PlaceVoii allowing existence of that graphical Envoii within Navigator. If a user drags the graphical Envoii to a different application (or if the Envoii is triggered to move spontaneously), such as Internet Explorer™, for example, a different PlaceVoii will be invoked to enable existence within that different application. Likewise, if an Envoii is dragged onto a desktop, a PlaceVoii appropriate for that desktop (in the particular operating system) will be downloaded (if necessary) and invoked.
MetaVoiis
A MetaVoii is an Envoii that exists at the highest level within a user's particular operating system. It detects relocation of a graphical Envoii to a new location and triggers the downloading and invocation of necessary PlaceVoiis to allow existence in different locations. In particular embodiments, a MetaVoii can trigger the loading and invocation of other MetaVoiis for different platforms, such as when an Envoii is relocated from a desktop computer to a PDA or to a different desktop computer on a network.
In particular embodiments according to the present invention, a graphical Envoii always has an ancestor that is a PlaceVoii, and a PlaceVoii always has a parent that is a MetaVoii. A MetaVoii generally has a parent that is a remote authoritative server for providing available PlaceVoiis, services, or other architectural entities. Generally, MetaVoiis and PlaceVoiis will not be associated with graphical objects directly and will operate transparently to users.
Connections
According to specific embodiments of the present invention, a connection specifies a relationship between two Envoiis. In this context, an Envoii can be understood as both an agent with which a user interacts and as anode in the Envoii architecture that supports that agent and may be transparent to a user. Nodes in the architecture provide formalized connection points for various Envoiis and supporting infrastructure.
Special Connections Examples
There are several types of connections that are peculiar to an Envoii system, according to specific embodiments. (1) Part/Part-of Connection is the structural glue that holds together “compositions.” It is a directed connection that says, “B is part-of A.” The intention here is to provide a formal method for identifying what gets “dragged along” when an Envoii is disconnected from its parent. (2) Kid: The Kid or “Child-of” connection is a special case of the Part/Part-of connection. It will be used to implement hierarchy within Envoii compositions. (3) Special Part Envoiis: All structural connections to an Envoii that are not Kids or Parents are Parts which facilitates service search procedures.
Visuals (Viz)
A Viz is a visual part of an Envoii that is normally perceivable by a user. A Viz subscribes to services such as: (1) Display manager; (2) Spatial Manager.
SkinVoiis (or Skins)
A skin (or SkinVoii) is a part that subscribes to events (user or otherwise) and, in general, operates on the properties of compositions or visuals. A skin can be understood as executable logic that provides interface functions for one or more Envoiis. According to specific embodiments of the invention, because a skin is apart, there is no limit to the number of skins that can be associated with a viz through its connected Envoii. Some of the services that are provided internally as methods to skins in some implementations can also be factored out to. Focus management is one example. Those skins that are interested in “focus” will obtain and connect to a special service called a “Focus Manager”. (This is an example of a Skin that is also a service).
SkinVoiis are responsible for interactions between the user and the Envoiis as well as interaction between Envoiis. They implement the behavior of Envoiis. In that sense, they mostly translate events into property value changes. SkinVoiis are parts. As such, a SkinVoii is directly attached to an Envoii and does not have parts or kids. A SkinVoii interacts with the rest of the system through its parent Envoii (service connections, access to property tree, etc.) When an Envoii is serialized (persistence, cut and paste, etc.), its SkinVoiis are also automatically serialized.
According to specific embodiments of the invention, SkinVoiis are event clients. Once attached to an Envoii, a SkinVoii will generally respond to a ConnectServices call (defined in the nvINode interface) by connecting to an event manager. A given SkinVoii may connect to other services to implement its specific behavior. (Note the spelling is some figures and examples may represent an older spelling using the final syllable “voy.” In this discussion, this is equivilant to spellings using the final syllable “voii,” which is the currently preferred spelling.) Example code follows: ConnectService(IID_nvIEvtMgr, (void**)&myIEvtMgr);
Once connected to an event manager, SkinVoiis register as event handlers for a set of events defined by a 32-bit mask, as follows:
myIEvtMgr->RegisterForEvent(m_EventMask,(nvISkin*)this);
The event manager will then call their PostEvent method anytime an event specified in the mask occurs. SkinVoiis also need to deregister and disconnect from the event manager when the DisconnectServices method is called. Example code follows:
Skin Voiis Modify Property Trees
When the SkinVoii is attached to its parent Envoii (for example, using SetParent) it is able to get access to the property tree of that Envoii. It is through that property tree that the SkinVoii can modify the state of its Envoii (or parts or kids, etc.).
Communication Done Through Skin Voii
According to specific embodiments of the present invention, the Envoii-to-Envoii communication model allows Envoiis to communicate even if they are not running in the same thread of execution (or even on the same machine). Because the software implementation according to specific embodiments of the invention is neither asynchronous nor re-entrant, communication between Envoiis needs to be synchronized by the event loop of the players. This means that messages between Envoiis should be carried by events, which indicates that communication between Envoiis has to be done through SkinVoiis. From the point of view of the SkinVoii, the communication manager is an auxiliary to the event manager which provides information that completes the communication event itself.
3. ServicesAccording to specific embodiments of the invention, the Envoii architecture is associated with a number of functions referred to as services. Services can be nested. This is especially desirable in the case of the display service, and implicit in the spatial service. All of the functions of an Envoii player (as discussed herein) can be cast as services. Services need not be local. This happens without further effort if one follows the chain of Envoiis up past the MetaVoii level at the local machine to the server level MetaVoii. Example services that may be associated with Envoiis according to specific embodiments of the invention include such things as the Communication Manager, Display Manager, Events Manager, Focus Manager, Memory Management, Tracking Manager, and Security Manager. Not all of these services will be included in each implementation.
4. Communication ManagerThe Communication manager allows a skin (or SkinVoii) to communicate with another skin. This skin can be located in the same PlaceVoii, in an other PlaceVoii within the same client or in a remote PlaceVoii. This section discusses the functionality implemented within the Communication Manager (ComMgr) component. The Communication manager (ComMgr) is instantiated at the level of the MetaVoii. In particular embodiments, it inherits from nvNode and is attached as a part of the MetaVoii. The ComMgr is not an Envoii, as it does not have any parent/child connection. The part mechanism allow each Envoii of a composition tree to access this service through the ConnectService method implemented in the nvNode interface. There will be an instance of ComMgr component for each instance of the MetaVoii within one client.
The communication manager allows skins to send and receive messages, with each message an instance of SkinMessage class. This class contain the message send between skin and routing/addressing information.
The nvIComMgr Interface
Each Envoii in the composition tree can request a connection from the ComMgr service (interface). An example embodiment source code is provided below.
The MetaVoii creates an instance of ComMgr, initializes it, and registers itself with this component (back pointer to the MetaVoii interface). The PlaceVoii received WM_CM_SKINMSG in its event loop. If the MsgId parameter is zero, the PlaceVoii forwards the message directly to its instance of ComMgr. If MsgId is not zero and the skin requested by this message is registered for event in this PlaceVoii, it will translate it as NV_SKINMSG event and send it to requested skin event loop.
The communication manager offers two methods to send messages from one skin to the other. If the skin initiator does not know the specific location of the destination skin, it uses a sendMsg with a scope of search. The scope can take the following value: SAME_PLACEVOII; LOCAL_PLACEVOII; REMOTE_PLACEVOII. If the skin knows the IP address and the port number of the PlaceVoii, it uses the second form of the sendMsg.
Sending Messages
When the send method is called, the ComMgr will ask its PlaceVoii if the requested skin is registered for events. If so, the ComMgr will store the message into a shared memory list, generate a MsgId and then using the MetaVoii interface will post a WM_CM_SKINMSG with a MsgId and a SkinId as a parameter. Upon reception of this message the PlaceVoii will convert the message into a NV_SKINMSG and send it to the event loop of the requested skin. On reception of the NV_SKINMSG the skin will proceed as follow:
Each ComMgr instantiates a Broadcaster object. A Broadcaster object creates a connectionless socket (such as using UDP) to a specific port (e.g. 1234). Each instance of the broadcaster has the ability to broadcast messages to a local area network. The first instance of the broadcaster will create a listening thread that will receive all the incoming broadcast messages. Upon reception of those messages, it will forward this message to all the PlaceVoiis within this client. In the current implementation the only message received on this port are IS_YOUR_SKIN message. Each instance of the ComMgr will create a thread which will be listening to pending connection. This thread is also in charge of checking all the connection object for incoming messages. Incoming messages are stored in the shared memory array and then the PlaceVoii is notified through its event loop.
5. Push ServerTracking is the ability for a player to gather data about how the user interacts with Envoiis, and then to send this information through tracking servers. For example, tracking how many times the user has clicked on an Envoii in a certain period of time, how long he left his mouse above it, etc. Tracking is completely anonymous. That means the system never makes any link between the data collected and the machine or the person who originated that data. Also, there is only tracking of Envoii deployed objects, no information is collected on the end user.
According to specific embodiments of the invention, a tracking system tracks events. It can use a special tracking SkinVoii that records specific events and forward them to a Tracking Manager at the player side. The Tracking Manager stores the data in a local file. Periodically, the Tracking Manager connects to the Tracking Server and sends reports.
Envoii Identity
For tracking information to be relevant, according to specific embodiments of the invention, the system provides a method for tracking the source Envoii. There are different possible levels of identification according to specific embodiments of the invention: (1) Envoii class level: for example, a graphical soft drink bottle Envoii. The system can track all clicks on all the bottle Envoiis in the world. (2) Envoii object level: e.g. to differentiate the clicks on two bottles living in the same place, for example a web page. (3) Player level: to differentiate a bottle on one user's desktop from a bottle on another user's desktop. (4) World level to identify a specific instance of a bottle Envoii throughout the world. To provide all these levels of identification, according to specific embodiments of the invention, a tracking system uses different kinds of IDs with different allocation methods. The way these IDs are allocated and propagated is linked to the Envoii life cycle, and Ids can be predefined properties with a special property type.
Player ID
To identify players throughout the world, according to specific embodiments of the invention, every player when installed registers with the Envoii server for a new Id. To do that, the invention delivers “virgin” players, i.e. players without an Id. During the first installation, the installer or bootstrap will try to connect to the Envoii server and request a new ID.
According to specific embodiments of the invention, player Ids can be stored with 4 bytes, which allows 2.sup.32 players in the world. A redirection mechanism on the server (e.g. in the CGI script) can be used to allow multiple servers, or to handle temporary server unavailability. The ID database can also count how many player Ids have been allocated per platform. On the end user machine, player Ids can be stored in different places, such as one of the player components (for example, the ORB (discussed herein) through a patch to DLLs). Ids may also be stored in the registry, though this location is liable to loss or modification by the user. Another option is the data directory, but this presents some of the same problems as the registry. According to specific embodiments of the invention, both in the registry and in the data directory is one choice.
Customer ID
According to specific embodiments of the invention, it is essential that Envoii customers (for Example, First Union Bank as described above) get tracking data on Envoiis deployed by those customers only. That means Envoiis must know to which customer they belong. According to specific embodiments of the invention, each customer has a customer ID (which can include a convenient number of bits, such as two bytes) and that id is stored by Envoii creation tools in the Envoii at compile time.
Class ID
According to specific embodiments of the invention, customers can differentiate tracking data for different types or classes of Envoiis (for example a bottle Envoii versus a logo Envoii). This is done using the Envoii Class ID. Envoii class Ids are allocated by the content creators that will keep track of the allocated Envoii class Ids.
Envoii Local ID
The Envoii local ID is a local number inside a place. Envoiis of the same class ID and same customer ID have different local Ids in the same place (web page, desktop, etc). This ID can be set at compile time for objects that will be created when the composition is created. For dynamically created Envoiis, local Ids can be allocated automatically, or with the help of specific scripting functions.
Normal Envoii ID
Normally, Envoiis do not need unique identification throughout the world. So the normal ID is simply the combination customer ID:Envoii class ID:local ID. In one example embodiment, the total size is 6 bytes. When a normal Envoii is copied, the Envoii ID is copied and a new local ID is allocated. When a normal Envoii is moved (Cut & Paste), the Envoii ID is unchanged. This will happen in the Copy/Paste and Drag & Drop operations.
Global Envoii ID
According to specific embodiments of the present invention, a Global Envoiis can be created with a unique identification throughout the world. This is accomplished using UUIDs (similar to those used for C++ class Ids). UUIDs are a 16 byte structure made from the following numbers: (1) The current date and time (2) A clock sequence and related persistent state to deal with retrograde motion of clocks (3) A forcibly incremented counter to deal with high-frequency allocations (4) The truly globally unique IEEE machine identifier, obtained from a network card (the implementation does not require a network card; if no network card is present, a machine identifier can be synthesized from highly variable machine states and stored persistently.) (The allocation algorithm is defined in Chapter 10 of “DEC/HP Network Computing Architecture Remote Procedure Call RunTime Extensions Specification Version OSF TX1.0.11” by Steven Miller, Jul. 23, 1992. This is part of the OSF DCE documentation.) When a normal Envoii is copied, a new Global Envoii ID needs to be allocated. When a global Envoii is moved (Cut & Paste), the Global Envoii ID is unchanged. This will happen in the Copy/Paste and Drag&Drop operations. Global Envoiis have to define which life cycle operations they allow. For instance, they could refuse to be copied. Thus, a Global ID can comprise the combination customer ID:Envoii class ID: UUID.
The Envoii ID can be stored according to specific embodiments using properties, which will provide a standard way to access the Envoii ID. According to specific embodiments of the invention, a special type of property (nvPropertyEnvoiiID) is in scripting, using a reserved property name called _EnvoiiID. Envoiis IDs generally are encrypted before serialization.
Tracking Reports
The tracking components on the player side collect all the tracking events sent by the SkinVoiis and store them into a local database. Once in a while, the local database is packed into tracking reports and sent over the Internet to the Tracking Server. These reports are called Reports of Transactions. In terms of implementation, reports of transaction need to be compact but extensible and secure. To verify data integrity, reports use checksums on each packet, plus one checksum on the report header.
Tracking SkinVoiis/Tracking Service
Tracking data is collected at the skin level. A SkinVoii inherits the capabilities of a base SkinVoii to do tracking. Tracking can be triggered per Envoii. The Envoii is tagged for tracking and needs to have a valid Envoii ID. The tracking flag is a reserved property called _trackable. This may be an integer working like a boolean: 1=tracking enabled, 0=tracking disabled. Other values are also possible. When a SkinVoii is connected to its Envoii, it reads the _EnvoiiID and _trackable properties. If the Envoii is trackable, the SkinVoii connects to the Tracking Manager service. On event reception (in PostEvent), the SkinVoii can decide to post a tracking event to the tracking service, using the TrackEvent( ) method of the nvITrackingMgr interface: void TrackEvent(nvPropertyEnvoiiID& Id, nvTrackingEvent& Event);
Each player (=each process) will dump its tracking data into a different file (Process Tracking Data). This is to avoid using any file locking mechanism that would be non-portable to MacOS. This file is a collection of fixed length records with records added at the end. According to specific embodiments of the invention, every 5 minutes or so, the tracking service will rename its tracking data file with a different extension (for example .rdy) and start dumping into a new file. The tracking service is waked up every minute by the PlaceVoii (using the PlaceVoii timer).
Files are created and parsed using the nvArchive object. That will ensure correct byte ordering, and provide a easy way to transfer data between the binary file and the structures in memory without bothering about byte alignment (which changes a lot across platforms). Raw data files are comprised of a header and a collection of fixed length records. A tool called rdydmp can be used to dump .rdy files into text files. According to specific embodiments of the invention, each record includes the following fields: Place Type, CustomerID, ClassID, LocalID, GlobalID, EventType, EventTimestamp, and EventData.
For processing a .rdy file according to specific embodiments of the invention, read every record, and update the RT database correspondingly. When the file has been processed, its deleted. Also, check for out of date raw data files, that is files that were never renamed into .rdy files, because the player was not terminated correctly.
In order to make the reporting mechanism as simple and reliable as possible, according to specific embodiments of the invention the invention uses an HTTP request because it is simple, reliable, allows a 2-way communication, and goes easily through firewalls. The report is sent in the data of the HTTP request and results are retrieved in the HTTP answer. On the Web server, there is a CGI script that saves the report to a file on the server. A separate reads this file and adds the corresponding data into a database (such as Oracle). This process will run asynchronously to the CGI script.
The Report Receiver will verify the integrity of the report. If the report is correct, a file name is allocated and the report is saved in that file. The receiver answers OK to the Tracking Manager. The Tracking Manager can delete the report from his local database. If the report is corrupted, the receiver logs an incorrect report incident (with some information on the report), and them answers ERROR to the Tracking Manager. In case of an error, either from Report Receiver or in the protocol with the Web Server, the Tracking Manager keeps the report for retry. According to specific embodiments of the invention, there may be a mechanism to discard incorrect reports in case they were corrupted on the player side and the Tracking Manager can check the integrity of the report with the same routine than the Reports Receiver.
8. Security ManagerSecurity is designed to ensure that the system as a whole is protected against malicious attacks as well as against internal mistakes. Important factors in assessing the need for security are: How much would a failure cost? How much does protecting against that failure cost? What is the perception of users? While there is always a way for a motivated and talented hacker to subvert a system and there is no way to reduce the talent level, the motivation level is pretty much under control of system design.
Components Integrity
Components are chunks of executable code that will be loaded and run by the players. As such they are capable of doing anything an executable is allowed to do on the clients' systems. This clearly pinpoints them as preferred targets for malicious attacks. In one design, according to specific embodiments of the invention, an authoritative server is the only source of components. This provides some added security.
With a highly controlled Envoii server as the only source of components for the users, this scheme cannot be subverted as long as the address from which components are to be fetched only comes from the Envoii server (as opposed to: comes from a page content). In addition to the single-source protection, components can carry a Verisign (or somesuch) identification to comfort users. Thus, the address used to fetch Components must be defined only by components. Components should use a system like Verisign Components safety.
Scripting
Scripts, just like components, contain executable code. However, there are two big differences: Scripts come from many sources. Scripts do not run in native code on a user's system but in the Envoii byte code inside a virtual machine. The first point makes security more difficult, and the second allows greater security measures to be taken. Because, according to specific embodiments of the invention, anybody can become a source for scripts simply by putting some Envoii content that refers to such script on a Web page. This means that, at least theoretically, anybody can run byte code inside of the Envoii virtual machine. Thus, the invention according to specific embodiments of the invention takes special care in defining the kind of access to the host machine's resources that is granted in the scripting language.
Tracking
Tracking also includes security against the emission by a tampered component of fake tracking reports. While the motivation for that is low unless there is some kind of reward attached to interacting with an Envoii and this kind of reward mechanism should only be implemented through communication, not tracking. To have any kind of significant impact on the overall reports, the fake reports would need to be very numerous. This kind of oddity should be very easy to spot on the tracking server side with basic statistical analysis and further the system can include a very low-level filter for impossible reports at the level of the tracking service just to make things a little more difficult for a hacker.
Communications
Communication channels established between Envoiis on users' systems and clients' severs might carry sensitive information so there is a risk (as for any transaction on the internet) of eavesdropping. User/client transactions therefore are encoded using the highest levels of encryption available (SSL or such).
9. Other ComponentsDisplay Manager
Executable code that enables display of Viz in different locations. Generally, there is one DM for each PlaceVoii.
Events Manager
Can either be per platform or in some cases additional events managers can exist in various local nodes on the tree.
Focus Manager
Not all Envoiis and their attached skins care about focus. Those that do register with a skin that is a focus manager.
Memory Manager
In specific embodiments, this can happen in a nested way, so that Envoiis get their own local memory manager which requests a block of memory from its (parent) memory manager. A system may need to override C++ constructors to use this system to have a clean way of cleaning up after disconnecting an Envoii.
Namespace Management
This manager provides support for global referencing to various nodes or leaves in an Envoii component-oriented architecture. This is related to Envoii identity. Each Envoii is assigned a locally and/or globally unique identifier. This identifier is used for: (1) Tracking—a compact key for generating reports (2) Finding mobile Envoiis. Identifier is used by remote ORB as search key. Mobile Envoiis need to report place changes to ORB hierarchy. (3) PlaceVoiis will need to know that they are in a special kind of place, like a Mac or PC or PDA. The services they request will need to be instantiated on the basis of their “place identity.” PlaceVoiis ask for services like: “MacPlace.EventServer.”
10. Service Discovery in Envoii ArchitectureWhen an Envoii is connected to its parent, or connected to any of its “grandparents”, it needs to discover and be wired up to a set of services, which will vary from Envoii to Envoii. One primary interface that Envoiis include is service discovery. In specific embodiments, the following method is used: (1) if you are an Envoii, ask your parts first, then ask your parent, (2) if you are a part, ask your Envoii. Thus requests for services flow upstream or laterally (to parts), but not downstream. This is the reason for the distinction between parts and kids. Because, according to specific embodiments of the present invention, services are nestable (multiple instance of the same service within the tree), the first service encountered is generally the one that is used, unless a particular service at a particular level is asked for by name (like: “ . . . / . . . /eventService” or “foo.eventService”.
11. User Desktop Experience Example ImplementationAccording to specific embodiments of the present invention, the desktop experience in windows is based on two different modules: bootdesk and PlaceVoii. Bootdesk is a bootstrap (an executable) module that creates a transparent window to be used as a desktop PlaceVoii. Bootdesk can be started on its own, or it will be started automatically when an Envoii is dropped on the desktop. When bootdesk is running, an icon appears in the taskbar. According to specific embodiments of the present invention, a right click on that icon gives access to a menu with different options to control bootdesk.
Transparent Window
Envoiis need a place to live. Basically, a place is a window. In a web browser, the window is provided by the browser. In the application bootstrap, the browser Envoii can be created in a regular window with a caption and a menu bar. For the desktop, according to specific embodiments of the present invention, a window is also needed, but in specific embodiments the window covers the whole screen, without obscuring the desktop. While one solution could be to use the desktop window itself, this does not work (under Windows) because one cannot subclass a window that was created in a different thread.
To meet this problem. the invention created a window that is made invisible and disabled in order not to disturb the desktop. If this window is created as a regular top-level window, there can be three problems: the window can be minimized automatically by the system, the window can appear above other applications, and a bootdesk application button will appear in the taskbar (at the bottom of the screen). To solve these problems, the window is created as a child of the program manager window, more precisely a child of the SysListView32 window (the window displaying the icons over the wallpaper).
HWIND hWndProgman=FindWindow(“Progman”,NULL);
HWND hWndShell=FindWindowEx(hWndProgman,NULL,“SHELLDLL_DefView”,NULL);
hWndShell=FindWindowEx(hWndShell,NULL,“SysListView32”,NULL);
Desktop PlaceVoii
How to draw in the created invisible and disabled window and how to handle the mouse inputs is the job of the desktop PlaceVoii. The desktop PlaceVoii uses the same subclassing technique as the plugin and the bootapp PlaceVoii. Subclassing of a window means that the original window procedure is replaced with a new procedure. The window procedure is a function called for every message received by the window. There are two major problems solved by this embodiment of the present invention. The first is how to draw Envoii objects on the desktop and have them move over it smoothly. The second is how to receive user inputs (mouse & keyboard) as well as repaint messages, given that the window is invisible.
Drawing on the Desktop
Drawing a sphere on the desktop is fairly straightforward. Using a regular alpha channel or a blue-screen, one can blit any image with a mask so that only the required pixels go to the screen. For example, that can be done using the SRCAND and SRCPAINT operators of the BitBlit or StretchBlit APIs. The harder problem to solve is how to repaint what was behind a sphere when it is moved. In other places, a PlaceVoii can use a special background mask that always renders itself first. On the desktop, there can be no background, otherwise the desktop would be obscured.
While the invention could ask the desktop to repaint itself when needed, e.g. when damages are pasted to the display manager, that is tricky for two reasons. The first is that there is no easy way to ask the desktop to paint itself without causing a lot of flickering. The second is that Envoiis are painted asynchronously, so it is almost impossible to repaint the desktop (with the icons and everything) in the same way.
Therefore, the invention has to paint the desktop itself. To do that, according to specific embodiments of the present invention, an algorithm saves the pixels of the desktop when and where needed, and then does the compositing between the Envoii rendering cache, the desktop cache and the alpha channel (see appendices for a detailed description).
If a window comes over the desktop PlaceVoii, the desktop PlaceVoii will be clipped automatically by the system, so any drawing in the rectangle of that window will not appear. As soon as that window disappears, because the system will send a WM_PAINT to the desktop windows, the desktop will repaint itself. When this is done, the PlaceVoii needs to repaint itself, That is done with system Hooks and is also a good moment to grab the pixels of the desktop, refresh the Envoii cache and paste the Envoii composition on screen.
Hooks are procedures that the system calls before or after executing certain system tasks like dispatching message. One can place a hook on a specific thread or on all threads. According to specific embodiments of the present invention, the invention places a hook on the Windows manager's application thread. For the repaint issue, hook the GetMessage( ) API to intercept all the WM_PAINT events on the desktop. When a WM_PAINT is intercepted, block the timer event in the PlaceVoii, so nothing is rendered while the desktop is repainting, and start another timer. As long as WM_PAINT is received, keep resetting that timer and merge all the update spans that are repainted. When the system goes idle (no more pending WM_PAINT), it means the desktop has finished repainting. The set timer will then come to an end, and the PlaceVoii will repaint on the accumulated spans. Before repainting, grab the new pixels on the screen. When that is complete, restart the main timer of the PlaceVoii so that animations can start again.
Receiving Mouse and Keyboard Input
The PlaceVoii window is invisible and disabled so it does not receive any mouse or keyboard input. This ensures that the desktop behaves as usual. To receive input on Envoiis, the invention employs a trick (under Windows) to create a hook that is a callback function that the system will call on every mouse input over a specific thread. In our case, set a hook on the program manager thread. To get the thread id of the program manager, do the following call:
hThread=GetWindowThreadProcessId(hShellWnd),NULL)
Then set a hook:
HookMouseProc=SetwindowsHookEx(WH_MOUSE, (HOOKPROC)MouseProc, hInst, hThread, NULL));
In the MouseProc( ) function, forward the mouse messages to the Envoiidesktop window using PostMessage( ). Generally, only the following messages are forwarded:
WM_LBUTTONDOWN WM_LBUTTONUP WM_RBUTTONDOWN WM_RBUTTONUP WM_MOUSEMOVEUse the same technique to set a hook on keyboard events to forward WM_KEYUP and WM_KEYDOWN messages. Both for the mouse and the keyboard, once the message is received in the PlaceVoii, check if some Envoii is interested in that message. For a mouse message, make sure the message is on a the shape of one of the Envoiis. For a keyboard message, check that one of the Envoii has the keyboard focus (=Envoii has a skin expecting keyboard events+Envoii was given focus by a click in his shape). If an Envoii wants the messages, set the capture of events to the PlaceVoii window, so that the desktop does not handle them in parallel.
Desktop Moving or Resizing
The desktop window used as a parent to the PlaceVoii window can be resized or moved in two different ways: (1) When the one of the dockable windows, like the task bar, is docked on a different side of the screen; (2) When the screen resolution is changed (display properties). In both situations, the PlaceVoii needs to be aware of the changes and resize the PlaceVoii window accordingly. The PlaceVoii window does not need to move the window to a different position because as a child of the program manager window, it is moved automatically. The PlaceVoii sets another hook on the window procedure of the program manager window. The set hook that will be called when the window procedure returns. If the message is WM_SIZE, resize the PlaceVoii window using the Windows API SetWindowPos( ) an then invalidate both the PlaceVoii window and the desktop window. SetWindowPos( ) will send a WM_SIZE message to the PlaceVoii window. The invention handles that message in the OnSize( ) method of nvPlaceVoiiOS.
Persistence
If the user logs off or shut down Windows, the desktop PlaceVoii should serialize its Envoiis in a file, and next time there is a log on (or boot), start bootdesk automatically and deserialize that file. There are three issues: how to serialize/deserialize the desktop, how to detect shutdown, how to start automatically.
Serialization/Deserialization
To serialize all the Envoiis on the desktop in a file, serialize the PlaceVoii in an archive and write that archive to a file. Because serialization is recursive, all the kids and parts of the PlaceVoii will get serialized, as well as all their descendants. Example code follows:
To deserialize, first read the file in an archive and call Placevoii->DeserializeProperties( ) This method will create all the Envoiis recursively. Before deserializing the PlaceVoii, existing Envoiis are deleted using PlaceVoii->RemoveAll( ).
Shutdown Detection
When Windows is shutting down or the user is logging off, the WM_QUERYENDSESSION message is sent to all top-level windows. Applications must return TRUE to allow the shutdown procedure to go on, FALSE to interrupt it. When everybody has returned, Windows sends a WM_ENDSESSION message to every top-level window again, to confirm or cancel the shutdown. If it is confirmed, this is the last message; after that, Windows terminates the application. Because the desktop PlaceVoii window is not a top-level window, it does not receive all these messages. To address this issue, the invention can create a hidden top-level window and use the WS_EX_TOOLWINDOW extended style to prevent the system from creating a button on the taskbar. CreateWindowEx(WS_EX_TOOLWINDOW, “EnvoiiDesktopTopWnd”, “ ”, WS_POPUP|WS_DISABLED, 0, 0, . . . ); In the window procedure the invention handles WM.sub.—QUERYENDSESSION and WM.sub.—ENDSESSION. On WM.sub.—QUERYENDSESSION, return TRUE to accept shutdown. On WM.sub.—ENDSESSION, serialize the desktop, set the system parameters to restart bootdesk automatically (see next paragraph) and destroy the PlaceVoii.
Automatic Restart (Autoboot)
If bootdesk is started with -boot in the command line, the desktop data file will be deserialized automatically. So call “bootdesk-boot” when Windows is booting or a new user is logging on. The invention does not use the usual Startup folder because it is language dependent and also it is easy for a user to remove the autoboot by mistake. Instead, add an entry in the registry at a key such as: HKEY_CURRENT USER\Software\Microsoft\Windows\CurrentVersion\Run. The entry is a command line to execute at boot time. It is a full path to bootdesk with the -boot option. That entry can be verified directly in the registry or use msconfig.exe (check the startup folder).
Forbid Multiple Instances of Bootdesk
To prevent Bootdesk from being started several times (thru icons, shortcuts, etc), bootdesk creates a named mutex at the beginning. If that mutex already exists, bootdesk displays a warning message and exits. Example code follows:
Desktop Refreshing Algorithm
According to specific embodiments of the present invention, the invention can be implemented using in part standards-based architectures and protocols for component-based programming. One such standard is known as CORBA (Common Object Request Broker Architecture), which has been developed by the 500+ member Object Management Group (OMG). Another such standard is the Component Object Model (COM), developed by Microsoft. According to specific embodiments of the present invention, the invention can be implemented in either of these standards-based environments. Some knowledge of these standards is presumed in the discussion below.
An Object Request Broker is a special component. It also provides objects, but in general the ORB does not build the objects itself. Instead, the ORB finds a component for the class needed and asks the component to allocate the object. In other words, the ORB is a kind of component manager. In a particular architecture according to specific embodiments of the invention, the ORB is a service. Any Envoii can request that service at any level. The ORB provides anonymous objects and named objects. In a particular OS, the ORB is implemented in a dll. It is loaded only once in memory but there is a separate data segment for each process. If the ORB needs to share data across processes, it can use classic shared memory techniques, though this is not required in all implementations. All objects are provided by components. An object class is identified by a unique ID called a CLSID (CLasS ID). CLSID according to specific embodiments of the invention is a random number that is intended to be unique in the world. According to specific embodiments of the invention, the invention uses the COM format for IDs (128 bits) and the COM id generator GUIDGEN.EXE. The ORB finds objects by name or by unique identifier. If persistent objects move from place to place, they need to register that fact with ORB. ORB exists at Meta level and generally will be nested. Generally, there will be atop-level (e.g. at Envoii.com) ORB for providing new components, primitives, etc., to various user sites.
nvorb is the project implementing the ORB (for VC++, load/src/nvorb/win/nvorb.dsp) Using the ORB is generally done in two steps. First, get an ORB interface to use ORB services. The following is sample code to do that:
With the ORB interface, one can call its methods. An example method is nCreateInstance( ) as provided in the example below. In this method, specify a class id, an interface id and a pointer to receive the interface. The ORB will find the component that implements clsid (provided it is in clsid.cfg), load it, create an instance of clsid, and ask this instance for an iid interface. This is returned in pObj.
The ORB makes the link between the clients and the components by using a correspondence table between CLSID and components. According to specific embodiments of the present invention, this table is built locally in a file according to a proprietary format. Whenever a new component is downloaded, the component registers its CLSIDs in that table. Later, the ORB is able to communicate with other ORBs to locate named objects (like CORBA ORBs do). The ORB also includes a loader to download components from a component server and resolve dependencies.
Loader
The ORB dll contains another object called the Loader. When a client asks the ORB to create an instance of an object, the ORB looks first into the CLSID table (the clsid.cfg file on disk). If that CLSID is unknown, or its component is missing (deleted by the user for instance), the ORB asks the loader to find that component. Now the loader needs an IP address for an ENVOII server or any other server that can talk with him. The CLSID table provides a list of such servers, in the [SERVERS] section. The communication takes place using the HTTP protocol so as to go through most firewalls without any change in the client configuration. On the server, there is an HTTP server running, which is generally already present because ENVOII servers will have HTTP to provide HTML pages and Envoiis. The Loader connects to the HTTP port of a web server and asks him to execute a CGI program called CGISTUB with the needed CLSID.
CGISTUB will now scan the CLSID table on the server for that CLSID. The CGISTUB can provide the web server with his response, which will be transmitted to the Loader. The response is a formatted block of text, with a mime header so that a browser can display it. Upon receipt of the reply, the Loader skips the mime header and parses the response. Zero, one or more components can be specified for download. For each one of them, the Loader checks that it does not have it on the client already. If not, it sends another query to the web server for a direct download of the component. The component is then saved to the ExeDir. When the loader returns, if the component that handles the CLSID has been found and downloaded correctly, the ORB loads that component (and its dependencies) and creates the object and finishes.
CLSID Repository
To find the component that implement a clsid, the ORB uses a CLSID repository. This repository can be implemented as a platform independent text file called clsid.cfg, located in the /common subdirectory of the Envoii data directory. This is an example of clsid.cfg
[CLSID]{40940D81-0772-11D4-8DAE-0050DAC05DOA}=skeleton.dll
As an example, for Windows: the Envoii data directory is specified in the registry at the following key: HKEY_LOCAL_MACHINE\Software\Envoii\Player\2.0\Data; the Envoii executable directory is specified in the registry at the following key: HKEY_LOCAL_MACHINE\Software\Envoii\Player\2.0\ExeDebug; other components are located at HKEY LOCAL MACHINE\Software\Envoii\Player\2.0\ExeRelease.
Resources
Resources are the data files needed by Envoiis to run correctly. For example, images, sounds, music files, etc. Those files can be located on the local disk, when designing new Envoiis, for instance, or they can be located on a web server, just like resources of a web page. Access to resources is generally transparent. To do so, the system uses the nvResource class. For each resource, the system: creates a nvResource object (through the ORB), gives it a name (that will be used in scripting) and a location (local file name or URL), loads the resource, gets a pointer on the data and the data size. Then, once done with the data, frees the resource. Example code is as follows:
Create an Object with the ORB Interface.
To use the ORB services, code first needs an ORB interface. Here is an example code to do that:
With an ORB interface, code can call its methods. Here specify a class id, an interface id and a pointer to receive the interface. The ORB will find the component that implements clsid, load it, create an instance of clsid, and ask this instance for an iid interface. Which is v/hu/you receive in pObj. An example method is nCreateInstance( ):
All interfaces inherit a base interface called nvIUnknown that implements reference counting and interface querying.
In specific embodiments, reference counting is not only defined but also implemented inside this class. Thus objects implementers do not need to do it themselves every time. So the only method that needs to be implemented is QueryInterface( ). In specific embodiments, there is also provided a class factory interface called nvIFactory, used to build class factories objets in components. The ORB used class factories to create object instances. The algorithm to create an instance is the following:
Player and Player Installer
In order for Envoii PIAs to exist on a client system according to specific embodiments of the invention, an Envoii player is placed on the client machine to handle Envoii PIAs. The Player, according to specific embodiments of the present invention, is made of several elements, the active components of which are discussed below. A set of elements according to specific example embodiments comprises: nvcore.dll, nvorb.dll, Envoii.dll, PlaceVoii.dll, MetaVoii.dll, commgr.dll, bootdesk.exe.
Bootstrap
According to specific embodiments of the invention, Bootstrap is begun the first time a user encounters Envoii content and may also start the first time an application is started. Bootstrap is platform and context dependant. It instantiates the ORB/Loader first and then executes the following boot sequence: (1) Instantiate a MetaVoii; (2) Instantiate a PlaceVoii to match the bootstrap context; (3) Tell the PlaceVoii to instantiate a root Envoii either from persistence data (for example from an HTML page) or from the MetaVoii's clipboard.
Installer
In an alternative implementation, a player is installed with user input. An installer is an executable module that, once downloaded on the client machine, will install the Envoii player in a specific directory, chosen by the user, and setup all the required parameters for the Player to run. This method of executable installation is typical for personal computer type-systems. The installer according to specific embodiments of the invention is used for 3 different tasks: (1) Do the very first install of an Envoii player; (2) Do an update of the Envoii player; and (3) Redo the setup in case the registry or the data directory is lost. The following files need to be in the same directory as the installer: all the components; an information file called Envoii-installer.inf (with the list of components); a table of CLSIDs called Envoii-installer.cfg.
Installing proceeds according to one example generally as follows:
1. Compare the source and destination directory (the source directory is where installer.exe started). If they are the same, assume that the user is trying to redo a setup after a loss of the registry or the data directory.
2. Create the data directory (if it does not already exist).
3. Create the executable directory (if it does not already exist).
4. Update the registry (release number, data dir, exe dir, player ID, proxy settings) and if a player ID is already in the registry, keep it.
5. Copy the Envoii plugin into Netscape and Internet Explorer directories if those directories are present.
6. Copy nvcore.dll and oldcore.dll to the Windows system directory.
7. If a re-setup, go to 12.
8. Copy installer files to the executable directory, to allow a re-setup if necessary.
9. Copy all the components specified in the installation file (Envoii-installer.inf) to the executable directory.
10. Create or update the CLSID table (clsid.cfg) and after that register all the existing components.
11. If installing a virgin player, attempt to get a player ID from the Envoii server.
12. Register plugin with IE and Netscape to handle .nvo files.
Player ID Allocation
According to specific embodiments of the invention, drag and drop is the ability provided to allow Envoii PIAs to be transported between “places” using an indication method, such as a pointing device such as a mouse. According to specific embodiments of the invention, a user can click on an Envoii, drag it on the screen while keeping the mouse button down, and finally release a pointer to drop the object somewhere else. The place where a user picks up an object is called the drop source. The place where a user drops the object is called the drop target. The object used between the drop source and the drop target is called the data object.
Typically, drop sources and targets are windows from one or more applications, though targest, especially, can be indications of other connected machines. Typically, applications involved in a drag and drop do not have to know each other. They can pass objects to each other provided they share some of the formats that these objects can take. These formats are usually called clipboard formats. There are some predefined formats for simple texts, bitmaps, files, GIFs, etc. When an object is dropped somewhere, if a user indicates to delete the original object, and this is called a move. If the original object is kept in place, it's called a copy. Move or Copy is called the drop effect. During the drag and drop operation, a user generally needs some feedback. For instance, he needs to know when dragging starts, or what object is being dragged. Also, when he moves his cursor over different windows, he needs to know which one is a potential drop target, and for which drop effect (copy or move). These operations must be performed by the drop source and the drop target. Also, the drop source and the potential drop targets need to know what is going on during the dragging. For instance, the drop source wants to know when the object is dropped, and with which drop effect. On the other side, the drop target needs to know which formats are available for the objects being dragged over him. Depending on the formats, he will or will not accept the drop.
To preserve a good interoperability between applications, as well as a coherent look and feel, the communication protocols between the drop source, the drop target and the data object, are usually defined by the system. For instance, drag and drop under Windows is done using OLE, which itself is based on COM. Applications implementing drag and drop often use system libraries.
Envoii Implementation of Drag&Drop
According to specific embodiments of the invention, because Envoiis perform cross-platform, the invention, rather than including its own drag and drop libraries wraps the system drag and drop features inside classes. According to specific embodiments of the invention, there are 4 of these classes: (1) nvDropSource implements a drop source; (2) nvDropTarget implements a drop target; (3) nvDataObject implements a data object; and (4) nvDataSource is a convenience class to ease the creation of the data object at the drop source. nvDropSource and nvDropTarget are base classes that are derived to implement some virtual methods. nvDataObject and nvDataSource are convenience classes that can be used as is. All these classes are defined in dragdrop.h, dragdropwin.h, dragdropwin.cpp (available in/common).
nvDropSource
nvDropSource is used to define the behavior of the drop source during dragging. The methods defined here are called repeatedly by the system during the operation. The drop source is the one who controls when a drag an drop ends (by a drop or a cancellation). Also, it can reflect the status of the dragging in any manner via the OnGiveFeedback( ) method. A drop source must be linked to a window, and this window must be registered with the system as a potential source of drag and drop. This is performed with the method Register( ). Example code follows:
nvDropTarget
nvDropTarget defines the behavior on the target side, so there is a little more to do than on the source. A drop target must be linked to a window, and this window must be registered with the system as a potential target of drag and drop. This is performed with the method Register( ). Example code follows:
nvDataObject
nvDataObject is a wrapper around the real data object. It provides two simple methods: one to retrieve the available formats for the object, and another to retrieve the object itself, in a specific format. The object is returned as a handle, which is a more general object descriptor than a pointer. Each implementation of nvDataObject will need to define its NVOBJECTDATA type (for Windows, it's a HGLOBAL). Same thing for clipboard formats with NVOBJECTFORMAT (for Windows it's a CLIPFORMAT). Example code follows:
nvDataSource
nvDataSource is another wrapper for a data object, but on the source side. One builds an object by calling CacheGlobalData( ) one or more times (once for each format you want to provide). Then you start the drag and drop by calling DoDradDrop( ). The SetClipboard( ) method is discussed later in this document.
Other Functions
Three more functions are provided to initialize an Envoii system drag and drop libraries (OLE) according to specific embodiments of the invention, to be called once in the application. DD_bRegisterClipboardFormat( ) is needed to register proprietary clipboard formats with the system (so that different applications will not use the same format identifier).
Using The Drag & Drop Classes
Using The Drag & Drop Classes is fairly straightforward. First, identify the windows that will be used for drop sources and targets. There will probably be a C++ object wrapping those windows where one will implement drag and drop. Then, derive nvDropTarget to implement the target side. The minimum done for enabling an Envoii is to override OnDragOver( ) and OnDrop( ). As an example, in OnDragOver( ) return the drop effect that will occur in case of a drop. To do that, check what formats are available in the data object. In this example, the only one known is CF_TEXT. If it is there, accept a move or a copy. Example code follows:
OnDrop( ) is called when the data object is dropped on your drop target. Here it is necessary to check the available formats again, and retrieve the data in a format that we can handle:
One needs to instantiate those two classes and register them, which means associate them with windows, and register the whole thing with the system drag and drop. Usually, one instantiates the drop source and the drop target in the object that wraps the window. Usually, a drag and drop is started on every click with the left mouse button. The DoDragDrop( ) method will take care not to start the drag and drop before a certain delay, or until you move the mouse (with the button down) outside a specific rectangle around the original click point. So, doing a simple click on a object, the drag does not start and the click is handled normally. Before starting the drag and drop operation, build a data object and wrap it in an instance of nvDataSource. Then start the drag by calling nvDataSource::DoDragDrop( ). Example code follows:
Interaction with System Clipboard
With an nvDataObject in which the invention can store whatever is necessary and pass it across applications, it is easy to use the system Clipboard and provide the usual Copy/Cut and Paste feature. Information on clipboard formats is available for various operating systems. Predefined values do not need to be registered.
Copy/Cut
To copy, the invention responds to a CTRL+C (under Windows), builds a data source as for drag and drop, and calls the bSetClipboard( ) method of nvDataSource. Now the object is on the clipboard. Any application can paste it using CTRL+V. The invention does not delete the nvDataSource object because the clipboard needs it. Deletion will occur automatically when the reference counting reach 0. To cut is exactly the same procedure, but deleting the local object. Example code follows:
Paste
To answer a CTRL+V, the invention builds an nvDataObject and calls the AttachClipboard( ) method. The invention can then ask for available formats with IsDataAvailable( ) and retrieve the data with getGlobalData( ). When the nvDataObject is deleted, the reference to the clipboard is released. Example code follows:
Drag and Drop Typedefs to Support Cross-Platform Operation
Because the upper level code must be cross-platform, some basic types of the drag and drop libraries are redefined. Example code follows:
According to specific embodiments of the present invention, the invention provides PIAs with a compositional model of content. This means that an Envoii web page, for example, is a hierarchical composition of various envoiis. The root of this hierarchy is the special envoii called PlaceVoii, because it represents a place that envoiis can exist. The children of this PlaceVoii are other envoiis.
Further, according to specific embodiments of the present invention, any envoii can dynamically become a child (part of) any other envoii. This capability is referred to herein as composability. Since each envoii in a composition is self-contained and autonomous, this enables a new model of software construction: (1) Users can personalize their experience and compose and aggregate useful envoiis in any configuration without loss of functionality. (2) Businesses can dynamically add new functionality, or upgrade old functionality in already deployed applications simply by sending new envoiis to their users. (3) Businesses can offer premium services, embodied as envoiis, to their users. Because of envoiis' built-in tracking capability, these services can be offered on an ASP or per-use basis. (4) Designers can rapidly develop new envoiis by “snapping together” already existing envoiis and customizing.
16. Further Implementation Examples According to Specific Embodiments (C++ Projects)Bootstraps
According to further embodiments, bootstraps can be used to automatically install player components when an Envoii is encountered. Bootstraps are small applications or plugins that can instantiate the Envoii player. There is usually one bootstrap for each kind of place. The first thing that a bootstrap does is to locate the ORB and load it. From there, it calls the ORB to instantiate different player objects that are needed to run (MetaVoii, PlaceVoii). Usually, the bootstrap transfers some data called boot data to the player. For example, a .nvx file.
Components
Components are dynamic link libraries exposing a standard set of functions that the ORB can call to instantiate objects. Each component provides a set of objects referenced through their class Ids, and used through interfaces. Only the ORB should load components.
Core Libraries
Core libraries are statically linked to other dlls, mainly components, but also applications. That means Core libraries are loaded automatically by the system, which implies that they generally must be accessible by the system before starting the player with a bootstrap.
Applications
Applications live on the client side or the server side. Some of them are launched by the client. Others can be launched by the player itself. On the server side, applications can be started by the administrator, or by any service like a web server, for example.
ServerVoys
ServerVoys are Envoii services running on the Envoii Server. Those objects are created and used by BootCGI on HTTP requests from clients. ServerVovs are used for tracking, playerID registration, component download, player update & Community servers.
Components
A component is a piece of executable code that provides objects. If one needs a FooBar object, there is provided a standard way to ask a component for that class of object (usually an exported function of a dll). One provides some kind of class identifier, some interface identifier and the component provides a pointer on an object. When the object is not needed any more, one of its methods is called to release it.
Objects and Interfaces
An object can be seen as a simple C++ object, that is an instantiation of a C++ class. Each class has a unique identifier (throughout the world) that provides for class identity. A component can implement several classes. An interface is a set of methods to access an object. The some interface can be used on different classes. A class can implement several interfaces. When a component is asked to create an object it does not return a pointer on the object but an interface to it. Most of the time it is the same, but sometimes not. In a particular embodiment, interfaces are C++ abstract classes. All interfaces inherit a base interface called nvIUnknown that implements reference counting and interface querying. In this example, this class can be defined in /common/nvcom.h.
Unique Identifiers
Unique identifiers are pseudo-random numbers that are unique in the world. According to specific embodiments of the present invention a COM-like format can be used for IDs (128 bits) and the COM ID generator GUIDGEN.EXE can be used to generate IDs. IDs are stored in a GUID structure taken from COM, such as the example below:
Class Factory
Components do not export functions to create objects directly. This is because it would not work for out-processor components (executables). So instead they provide class objects, also called factories. These factories are C++ objects with a default interface called nvIFactory (defined in nvcom.h). Their method nCreateInstance( ) is the one who provides the objects themselves. There is one factory for each class implemented by a component. That is why the method nCreateInstance( ) shown below does not have a class id parameter.
An example of the steps to create a FooBar object are the following: (1) ask the component for a FooBar factory; (2) ask the FooBar factory object to create an instance of FooBar and return an interface on it; (3) release the factory object. FooBar can have multiple interfaces. In step 2, specify one of them. If later it another interface is desired on the same FooBar object, call the QueryInterface( ) method of the FooBar object itself. Generally, it is necessary to release ALL the interfaces that you get, either with nCreateInstance( ) or QueryInterface( ).
//query interface int nvIUnknown::QueryInterface(const IID & iid, void **ppObj);
Building a Component
/src/skeleton is an example of a component. It implements two classes: Skeleton and Skeleton2, and two interfaces: ISkeleton and ISkeletonFoo. The source are organized as follow:
nvObject—This is the root class. All classes inherit from it. It gives the ability to have a central access point to all objects in the code. For example, allowing the code to modify the way objects memory allocation is handled, or provide polymorphic persistence mechanisms. Indexed list—Can use resizable indexed lists as they are faster to traverse and lighter in memory.
18. COM-Like Example ArchitectureBelow is a discussion of a COM-like example architecture according to specific embodiments of the present invention. This discussion provides a very basic example of how to define classes that will work in Envoii components according to specific embodiments of the present invention and be usable by the ORB and provides further information regarding processing behind the scenes. This document is a wrapper for a C++ file. According to specific embodiments of the invention, this particular architecture defines three different classes: class Node; class Envoii, which inherits class Node; class PlaceVoii, which inherits class Envoii, and for each class, the equivalent interface: class Inode; class IEnvoii; class IPlaceVoii. Two functions to create the right objects are created: GetPlaceVoii and GetEnvoii. In reality, this is done by the ORB which gets factories from the components and ask those factories to create the objects.
Important Rules
To design classes and interfaces that will work in this example COM-like, the following rules should be respected:
1. All interface inherits class nvIUnknown, as stipulated in COM. This is to allow anybody who has an interface on an object (=a pointer) to query another interface using the QueryInterface method of nvIUnknown.
2. Because PlaceVoii and Envoii objects inherits the Node object, and because there must be only one reference counter per object, reference counting is implemented only once, in the base class Node.
3. Because of inheritance between interfaces, define the AddRef and Release method in every class. They can either bounce on the base class, or do the same job.
4. For the same reason, define QueryInterface for every class. Each QueryInterface should only return a pointer for its own interfaces and call the base class QueryInterface for the others. This is to allow the this pointer to be casted correctly. (=offsetted to the right virtual table).
5. A QueryInterface that succeeds should do an AddRef( ).
6. All methods of all interfaces must be virtual. Destructors MUST be virtual too.
Example Code (oop.cpp)
The invention also may be embodied in whole or in part within the circuitry of an application specific integrated circuit (ASIC) or a programmable logic device (PLD). In such a case, the invention may be embodied in a computer understandable descriptor language, which may be used to create an ASIC, or PLD that operates as herein described.
21. Other EmbodimentsThe invention has now been described with reference to specific embodiments. Other embodiments will be apparent to those of skill in the art. In particular, a viewer digital information appliance has generally been illustrated as a personal computer. However, the digital computing device is meant to be any information appliance for interacting with a remote data application, and could include such devices as a digitally enabled television, cell phone, personal digital assistant, etc.
In addition, channels have been described primarily as traditional network connections, with the appropriate corresponding hardware. However, channels are meant to be any channels capable of carrying data, including wireless channels, optical channels, and electrical channels.
It is understood that the examples and embodiments described herein are for illustrative purposes and that various modifications or changes in light thereof will be suggested by the teachings herein to persons skilled in the art and are to be included within the spirit and purview of this application and scope of the claims.
All publications, patents, and patent applications cited herein or filed with this application, including any references filed as part of an Information Disclosure Statement, are incorporated by reference in their entirety. Conclusion
The invention has now been explained with regard to specific embodiments. Variations on these embodiments and other embodiments will be apparent to those of skill in the art. The invention therefore should not be limited except as provided in the attached claims. It is understood that the examples and embodiments described herein are for illustrative purposes only and that various modifications or changes in light thereof will be suggested to persons skilled in the art and are to be included within the spirit and purview of this application and scope of the appended claims. All publications, patents, and patent applications cited herein are hereby incorporated by reference in their entirety for all purposes.
Claims
1-20. (canceled)
21. A system for providing a portable information agent comprising:
- a data processing system that can present a graphical representation of an original portable information agent as part of a first composition accessed from a data memory by an initial application, where the original portable information agent can comprise a software object, can have state, and can have one or more external connections;
- a data processing system that can replicate the portable information agent; and
- a data processing system that can relocate the replicated portable information agent such that the replicated portable information agent can become part of a second composition, where the second composition can provide a persistent link between the original portable information agent and the replicated portable information agent.
22. The system according to claim 21 wherein the replicated portable information agent, once relocated, can persist and maintain the state after termination of the initial application.
23. The system according to claim 21 wherein the initial application can comprise an email client and the first composition can be part of an email message.
24. The system according to claim 21 wherein the initial application can comprise a desktop provided by an operating system.
25. The system according to claim 21 wherein the data processing system for relocation can allow relocation to be repeatable from a current location to a plurality of additional locations.
26. The system according to claim 24 wherein the desktop provided by the operating system can comprise an interface of a platform, and the platform can be selected from a group consisting of: a Windows PC, a Macintosh PC, a Unix-type operating system, a set-top box, a wireless logic appliance, an Internet appliance, a personal digital assistant, or another device connected to a network.
27. The system according to claim 21 wherein the relocation is to a location selected from a group consisting of: a desktop provided by an operating system, a different application, and a different computer platform with a different operating system.
28. The system according to claim 21 wherein the replicated portable information agent can include one or more user interface components and wherein the user interface components can be preserved after relocation.
29. The system according to claim 21 wherein the replicated portable information agent can include one or more connections to one or more external entities and wherein the connections can be preserved after relocation.
30. The system according to claim 21 wherein the data processing system that can replicate can allow a user to select and drag the graphical representation.
31. The system according to claim 21 wherein the data processing system that can replicate can allow a user to discontinuously select the graphical representation and place the graphical representation in a new location.
32. The system according to claim 21 wherein the data processing system that can replicate can allow the graphical representation to move to a new location without an action by a user.
33. The system according to claim 29 wherein the one or more external entities can be selected from s group consisting of: web servers, other applications, background processes, and other remote processes.
34. A system for interacting with a portable information agent comprising:
- a data processing system for providing an original portable information agent object, using as part of a first composition accessed by an initial application from a data memory, the original portable information agent object comprising a software object, having state, and having one or more external connections;
- the data processing system for allowing a user to select the original portable information agent object and replicate the original portable information agent object remote from the initial application, and upon selecting and replicating, to determine if sufficient infrastructure logic exists to allow the replicated portable information agent object to operate in a new location; and
- the data processing system can thereafter move the replicated portable information agent object to the new location such that the replicated portable information agent object can become part of a second composition.
35. The system according to claim 34 wherein the new location can be selected from a group consisting of: a desktop provided by an operating system, a different application, and a different computer platform with a different operating system.
36. The system according to claim 34 wherein the replicated portable information agent object can be associated with a graphic object constructed from presentation primitives that allow user interaction with the graphic object.
37. The system according to claim 34 wherein the replicated portable information agent object can include logic allowing communication with a data server for tracking user interaction.
38. The system according to claim 37 wherein the replicated portable information agent object and the infrastructure can allow communication with a data server for tracking user interaction after the replicated portable information agent object has been moved to the new location.
39. A system for providing a portable information agent comprising:
- an original portable information agent stored in electronic data memory that is part of a first composition accessed by an application operating on a first processor, the original portable information agent comprising a software object, having state, and having one or more external connections; and
- a relocated portable information agent that is part of a second composition accessed by an application operating on a second processor, the relocated portable information agent comprising a software object, having state, and having one or more external connections including a persistent connection to the original portable information agent.
40. The system of claim 39 further comprising a plurality of relocated portable information agents that are each part of one of a corresponding plurality of compositions, each accessed by one of a corresponding plurality of applications, each operating on one of a corresponding plurality of processors, each of the relocated portable information agents comprising a software object, having state, and having one or more external connections including a persistent connection to the original portable information agent.
Type: Application
Filed: Feb 21, 2012
Publication Date: Jun 14, 2012
Inventors: Michael Tolson (Corte Madera, CA), Jean-Christophe Arnulfo (Paris), Brice Rive (Paris)
Application Number: 13/401,717
