Electrical computers and digital processing systems: multicomput – Remote data accessing
Reexamination Certificate
1999-01-13
2001-08-07
Dinh, Dung C. (Department: 2757)
Electrical computers and digital processing systems: multicomput
Remote data accessing
C707S793000
Reexamination Certificate
active
06272536
ABSTRACT:
BACKGROUND OF THE INVENTION
This invention relates to the distribution of software over a network. More particularly, this invention relates to the broadcasting of code and data, and updates thereto, to a plurality of subscribers.
In large scale networks such as the Internet, or Intranets within businesses, the distribution of software applications is often a manual and laborious process which requires the correct use of such program tools such as Rp, tar, compress, uudecode, and zip. The variety of platforms and tools, and the complexity of the installation procedures make this manner of distribution a complex and costly operation. Software installation is therefore frequently performed by specially trained system administrators rather than end-users.
The Internet has significantly accelerated the release schedule of applications significantly. Software is released more frequently and in smaller increments, and as a result many more installations have to be performed, resulting in more work for the system administrator. This multitude of releases can cause versioning problems when a new piece of installed software becomes incompatible with some previously installed software. As these updates occur more often, it is desirable to automate this update process.
A browser is a computer program for accessing the Internet via the World Wide Web, using the HTTP protocol. Browser plug-ins allow the user to extend the browser so that it can incorporate new functionality. Plug-ins are often very hard to install because they are platform dependent, and not secure because they are implemented in low level languages such as C or C++. To make plug-ins secure the browser needs to implement some form of authentication algorithm such as those based on the RSA algorithm.
The Java programming language and the introduction of Java applets has made it possible to run the same software program in a secure manner on many different platforms, thus enabling the wide distribution of such programs over a heterogeneous network such as the Internet. With Java applets it has also become possible to automatically launch small Java programs from a World Wide Web browser which eliminates a lot of the installation headache.
When Java applets are used as applications, the user is required to use a browser to navigate to the HTML page containing the desired applet. Once the applet is running it is usually constrained to the HTML page in which it is embedded, and the applet may be terminated prematurely when the user visits a new HTML page.
Further, Java applets have several restrictions which prevent them from scaling to larger applications. One problem is that the download times are too long because each Java class is loaded using a separate HTTP connection, and making each new connection often takes more time than the actual data transfer. Also, Java applets have to be reloaded from their source each time they are used, there is no mechanism for persistence other than HTTP caching. HTTP caching has the drawback that it is too low level, which causes versioning problems because it may mix old Java classes with newer Java classes. It is usually impossible to flush an applet from an HTTP cache, because it is not possible to know which files in the cache belong to the applet that needs to be flushed.
Because Java applets are reloaded for each use, and because they usually consist of many parts, they can significantly increase the number of server accesses and thus significantly increase the server load. As a result most high volume web-sites cannot afford to put Java applets on their HTML pages.
A Java applet generally cannot be used when the client computer is disconnected from the network. If the user wants to use an applet after disconnecting the network, it is first necessary to use all the features of the applet to populate the HTTP cache. However, if the user ventures into a previously unexplored part of the applet once disconnected, the applet will be unable to proceed and a fatal error will result. This is a major drawback of caching strategies because disconnected use is important for the next generation of portable Internet devices.
Another drawback of HTTP is that ongoing transactions can often be corrupted when new code and data is installed on the server. This is not a fatal problem when it happens to an HTML page, but corrupting a request can be catastrophic for an application. HTTP simply is not an appropriate mechanism for distributing code.
Some applications need the ability to monitor their use by the end-user. An example is advertisement tracking where it is necessary to keep an accurate count of how often an ad is shown. Currently advertisements are tracked by counting the number of hits on the HTML page containing the ad, which means that the page cannot be cached because this would affect the hit-rate. The page is therefore marked as not-cacheable, and the result is fewer cache hits and increased server load.
Personalized web-sites and personalized applications are also becoming very popular. Because of security constraints it is very hard to create a personalized applet. The reason is that the applet can't save state to the local disk and therefore all of the applet's persistent state must be stored in a database on the server, which further increases the server load.
What is needed is a way to distribute applications in a user-friendly, scaleable, secure, and seamless way, which enables the monitoring and personalization of applications, across a network such as the Internet.
SUMMARY OF THE INVENTION
In summary, the invention consists of a system and method for the large scale distribution of application code and data. The system consists of a client-side component, which the user uses to download applications, as well as a server-side component, which is used by a content provider or developer to make applications available for distribution.
The system also allows for the automatic updating, personalization, and usage monitoring of applications. In addition it is possible to use the application even when the host computer is not always connected to the network.
In describing this invention the following terminology from the broadcasting and publishing world is used:
Channel—A named application consisting of code and data which can be subscribed to by an end-user. Once delivered to the end-user a channel behaves much like a normal application.
Tuner—This is a client application which is used by the end-user to subscribe, download, execute, and administer channels. The tuner is responsible for the automatically updating of subscribed channels.
Transmitter—This is the server component of the system. It is responsible for delivering channels and channel updates to clients. The transmitter allows channels to be updated without interruption of service.
The end-user uses the tuner to subscribe to channels. When the user first subscribes to a channel a request is made to the transmitter hosting the channel. As a result of this request, the code and data for the channel is downloaded and stored in a local persistent storage, usually a hard disk. Once the entire channel is downloaded and installed on the local hard disk it can be executed as a normal application. The user can access the channel many times without requiring further network traffic.
At regular intervals the tuner can re-establish a connection to the transmitter from which the channel was obtained and check if the channel has changed. If changes are detected, these changes are downloaded and installed automatically.
REFERENCES:
patent: 4432057 (1984-02-01), Daniell et al.
patent: 4468728 (1984-08-01), Wang
patent: 4558413 (1985-12-01), Schmidt et al.
patent: 4611272 (1986-09-01), Lomet
patent: 4658093 (1987-04-01), Hellman
patent: 4714992 (1987-12-01), Gladney et al.
patent: 4714996 (1987-12-01), Gladney et al.
patent: 4866611 (1989-09-01), Cree et al.
patent: 4875159 (1989-10-01), Cary et al.
patent: 4897781 (1990-01-01), Chang et al.
patent: 4974149 (1990-11-01), Valenti
patent: 5113519 (1992-05-01), Johnson et
Payne Jonathan
Shaio Sami
van Hoff Arthur A
Dinh Dung C.
Gotlieb Charles E.
Marimba, Inc.
LandOfFree
System and method for the distribution of code and data does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for the distribution of code and data, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for the distribution of code and data will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2463286