Executing applications at appropriate trust levels (09-Mar-2010)

CROSS REFERENCE TO RELATED PATENT APPLICATION

This is a divisional of and priority is claimed to co-pending U.S. patent application having Ser. No. 10/857,689 and a filing date of May 27, 2004 for EXECUTING APPLICATIONS AT APPROPRIATE TRUST LEVELS of Stott, et al. This co-pending United States Patent Application is commonly assigned herewith and is hereby incorporated herein by reference for all that it discloses.

TECHNICAL FIELD

This invention relates to executing applications.

BACKGROUND

Executing applications deployed from remote sources can be dangerous. Applications from remote sources may contain malicious code like worms or viruses that can damage or misuse a user's computer or information.

To partially combat this problem, typical Internet browsers can execute an application published to a remote location (e.g., an Internet domain) with a trust level predetermined for that location. Applications executed at a high trust level are permitted to perform riskier operations than those executed at a low trust level. Trust levels used by these Internet browsers are typically set prior to running the application based on how trustworthy the remote location is deemed to be. To execute applications with these Internet browsers, however, a user typically must have access to the remote location, such as via the Internet.

If a user wants to execute an application published to a remote location for later use when he or she will not have remote access, the user can save the application onto his or her local machine. The user can then later execute the application when he or she does not have remote access. There is a significant danger in doing so, however. The application may not execute at an appropriate trust level when executed from the user's local machine. This is because applications loaded from a local machine typically execute with a higher trust that is assigned to the local machine.

Similarly, if a user wants to execute an application that is not published to—but does originate from—a remote location, the user can save the application onto his or her local machine. The user can then execute the application but it may execute at an inappropriate trust level. One common example of this is when applications are received via email or floppy disk. While the user can run these applications, to do so the user typically saves the application to his or her local machine, often implicitly granting the application a higher trust level than it deserves.

In these and other cases where an application is received from a remote source and saved locally, the trust level at which the application is executed can be too high or too low. This is because many computer systems assume a particular level of trust (usually too high) for applications cached or executed from a local source. This potentially endangers a user's computer and, importantly, personal or corporate information.

Assume, for example, that Joe emails Jane an application and Jane saves the application onto her local machine. By so doing Jane can execute the application from her local machine. When Jane executes the application from her local machine, however, her computer typically assumes a trust level based on the location from which the application was executed (locally), which is often inappropriate. If the application contains malicious code, when Jane executes the application from her local machine it may damage her computer, steal information, and the like.

Similarly, if Jane saves locally an application from a website and later executes it, the application is typically granted too high a trust level. If it is granted too high a trust level the application is executed at the higher, inappropriate trust level, thereby endangering her computer and its information.

Further, even if the application Jane runs is not given too high a trust level, but just a different trust level than that at which it will optimally execute, the application may perform inconsistently or otherwise operate poorly.

Thus, typical trust levels granted in executing applications locally that originate from remote sources are often too high or too low, either potentially endangering a user's computer or sacrificing consistent and/or robust operation of the application.

SUMMARY

This patent application describes systems and methods (“tools”) enabling execution of applications at appropriate trust levels. These tools can determine appropriate trust levels by comparing applications' permitted trust levels with their requested trust levels.

These tools can determine and embed requested trust levels into applications. The requested trust levels can permit or minimally permit operations capable of being performed by the applications.

To determine permitted trust levels, these tools can compare applications' execution locations with their published locations. The applications can then be executed at these permitted trust levels or at lower trust levels if the applications request lower trust levels. These tools can also disallow execution of applications that will not run safely and robustly, such as when an application requests a higher trust level than is permitted.

These tools also allow applications to be executed at appropriate trust levels when those applications are received from remote sources, such as through email or floppy disks. Regardless of from where applications are received, the tools can enable execution of these applications at appropriate trust levels.

Also, these tools can execute applications at a restricted trust level. Applications executed at this restricted trust level can be prohibited from performing operations capable of endangering a user's computer or information.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 illustrates an exemplary architecture having security tools.

FIG. 2 sets forth a flow diagram of an exemplary process for building requested trust levels.

FIG. 3 illustrates an exemplary table of trust levels.

FIG. 4 sets forth a flow diagram of an exemplary process for executing an application at an appropriate trust level.

FIG. 5 illustrates an exemplary table showing permitted trust levels.

FIG. 6 illustrates an exemplary table showing appropriate, permitted, and requested trust levels.

FIG. 7 sets forth a flow diagram of an exemplary process for executing or preparing for execution an application at a restricted trust level.

The same numbers are used throughout the disclosure and figures to reference like components and features.

DETAILED DESCRIPTION

Overview

This patent application describes systems and methods (“tools”) for secure execution of applications. Some of these tools can determine operations executable by an application that may damage a computer. Based on this determination or otherwise, these tools can build requested trust levels for applications.

Applications having a requested trust level can be sent from remote locations and received by tools located elsewhere, such as at a user's computer. The tools at the user's computer can determine whether or not to execute these received applications at their requested trust levels. In doing so, these tools can determine permitted trust levels for these applications as well as their requested trust levels. These tools can determine these permitted trust levels based on from where the applications are cached or executed and various information embedded into the application, for instance. In part by comparing the permitted trust levels with the requested trust levels, these tools can execute applications at an appropriate trust level, if one exists.

If, for instance, an application requests a higher trust level than the tools have determined to be permissible, the tools may not execute the application. Here, the tools potentially protect a user's computer from an application that may contain malicious code. Also for instance, if an application requests a trust level identical to that which the tools have determined permissible, the tools can execute the application at the requested trust level. Further, if an application requests a lower trust level than that determined to be permitted, the tools can execute the application at the lower, requested trust level. By so doing, the tools can execute applications at an appropriate trust level.

In cases where a very low trust level is appropriate for an application, the tools can execute the application at a restricted trust level. This trust level enables safe execution of applications that may not be trustworthy.

Exemplary Architecture

Referring to FIG. 1, an exemplary architecture 100 is shown having a computing device 102. The computing device 102 is shown capable of communicating with a remote location 104 through a communication network 106 or physical media 108. The remote location 104 can comprise locations at which accessible information is stored, such as computing devices or an Internet domain. The communication network 106 comprises devices or manners by which the computer 102 can send information to, or access information at, the remote location 104. The communication network 106 can comprise, for instance, a global Internet or an intranet. The computing device 102 can, for instance, send applications to, and receive applications from, the remote location 104 through email via the communication network 106. Applications can also be sent and received through physical media 108, such as floppy disks.

The computing device 102 is shown having access to or comprising a processor 110, an operating system 112, a memory 114, and security tools 116. The processor 110 and the operating system 112 are well known and so are not discussed here. The memory 114 can comprise volatile memory and/or non-volatile memory. The memory 114 is shown with a cache 118 and a local memory 120. To aid in discussing various embodiments of the tools 116, the tools 116 are shown having a trust-level builder 122 and a runtime 124. Also to aid in discussing various embodiments, an exemplary application 126 is shown. This application 126 can comprise any compilation of executable code, such as a form template or a word-processing document having a macro. This application 126 can originate, be executed from, and/or be cached from various locations, such as the local memory 120 or the remote location 104.

This architecture 100 and its components are shown to aid in discussing, but are not intended to limit the applicability of, the security tools 116. Other well-known computing systems, environments, and/or configurations that may be suitable for use with the tools 116 comprise, for example, personal computers, server computers, hand-held or laptop devices, multiprocessor systems, microprocessor-based systems, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.

The tools 116 may be described in the general context of, or implemented with, computer-executable instructions, such as program modules, executed by one or more computers or other devices. Generally, program modules include routines, programs, objects, components, data structures and etc. that perform particular tasks or implement particular abstract data types. The functionality of the program modules may be combined or distributed in various embodiments, including those described below.

These computer-executable instructions can comprise computer-readable media. The computing device 102 can, for instance, comprise computer-readable media, which can be accessed by the tools 116. Computer-readable media can comprise, for example, computer storage media and communication media. Computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information. This stored information can comprise computer-readable instructions, data structures, program modules, and other data. Computer storage media comprise, by way of example, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVDs) or other optical storage, magnetic media storage devices, or any other medium that can be used to store the desired information and that can be accessed by the tools 116. Communication media typically embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal (e.g., a carrier wave or other transport mechanism) and includes any information delivery media. The term “modulated data signal” means a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. Communication media can comprise, for example, wired media, such as a wired network or direct-wired connection, and wireless media, such as acoustic, RF, infrared, and other wireless media. Combinations of any of the above can also be included within the scope of computer-readable media.

Building Requested Trust Levels

Referring to FIG. 2, an exemplary process 200 for building requested trust levels is shown. The process 200 is illustrated as a series of blocks representing individual operations or acts performed by the tools 116 and/or the builder 122. This and other processes described herein may be implemented in any suitable hardware, software, firmware, or combination thereof. In the case of software and firmware, these processes represent sets of operations implemented as computer-executable instructions.

The tools 116, through the builder 122, can build requested trust levels for applications and, through the runtime 124, execute these or other applications at an appropriate trust level. The applications, such as the application 126, executed by the runtime 124 may be those having requested trust levels built by the builder 122 or may be received from the remote source 104. Applications received from the remote source 104 may have requested trust levels built by some other builder 122, and so may or may not be trustworthy. For purposes of this description of the process 200, the builder 122 prepares the application 126 for later use, such as by another user at a remote location, by building a requested trust level for that application 126.

Determining Potentially Dangerous Operations Performable By an Application

At block 202, potentially dangerous operations performable by an application are determined. In an ongoing embodiment, the builder 122 analyzes the application 126 to determine what types of operations it can perform that can potentially harm the computing device 102 or compromise its information. The builder 122 can analyze the application 126 by scanning its constituent parts to find links, data sources, web services, and other pieces of code that can indicate a potential compromise to security.

The builder 122 can, for instance, find universal resource locators (URLs) indicating that the application 126 may attempt to communicate with remote locations, such as the remote location 104. How the application 126 intends to communicate can also be ascertained by analyzing code associated with a URL. This associated code can, for example, look information up from or send information to data sources like a database or an Internet site. Similarly, the builder 122 can find code that accesses personal information of the user (such as information stored in the memory 114) and sends it out, such as the user's credit card information sent to a bank website. The builder 122 can also determine that no code of the application 126 can compromise security.

At block 204, a minimum trust level is determined. In the ongoing embodiment, the builder 122 determines a minimum trust level at which the application 126 is permitted to perform the potentially dangerous operations determined at the block 202.

Referring to FIG. 3, an exemplary table 300 of trust levels is shown. In this embodiment, these trust levels comprise three levels: full trust level 302; location-dependent trust level 304; and restricted trust level 306. Full trust level 302 permits any operation by the application 126. Location-dependent trust level 304 permits operations not requiring full trust 302 by requiring at least one operation that can potentially compromise security. Restricted trust 306 does not permit any potentially security-compromising operations.

If, for example, the builder 122 determines that the application 126 can access, add, alter, or delete information from the memory 114, the builder 122 determines the minimum trust level to be the full trust level 302. If the builder 122 determines that the application 126 can access information from a website, the builder 122 determines the minimum trust level to be the location-dependent trust level 304. If the builder 122 determines that it cannot access any information other than the information it creates, the builder determines the minimum trust level to be the restricted trust level 306.

Setting a Requested Trust Level

At block 206, a requested trust level is set. This requested trust level can be set by a user, such as by the user manually choosing the trust level. In the ongoing embodiment, the requested trust level is the minimum trust level determined by the builder 122 at the block 204.

At block 208, the requested trust level can be embedded into an application. In the ongoing embodiment, the builder 122 embeds the requested trust level into the application 126. The builder 122 can do so by adding code into a configuration setting or another appropriate location of the application 126. If the application 126 comprises eXtensible Markup Language (XML), the builder 122 can add the XML attributes set forth for the trust levels in the table 300 of FIG. 3.

Referring to FIG. 3, the builder 122 adds the attributes shown in the table 300 to the application 126, based on the requested trust level. The builder 122 can add, for instance, a requested full trust level 302 to the application with the full-trust attribute 308. Here the attribute 308 is: “requireFullTrust=yes”. For a requested location-dependent trust level 304, the builder 122 can add the location-dependent attribute 310. Here the attribute 310 is: “trustLevel=Domain”. In some embodiments, the attribute 310 can also be “trustLevel=”, which can be assumed by the runtime 124 to equate to “trustLevel=Domain” but with “Domain” being a different location than an execution location, discussed below. For a requested restricted trust level 306, the builder 122 can add the restricted attribute 312. Here the attribute 312 is: “trustLevel=Restricted.”

At block 210, an application is published to a location. In the ongoing embodiment, if the trust level requested is the location-dependent trust level 304, the builder 122 embeds this published location (e.g., a dependent location URL) information into the application 126. The location-dependent trust level 304 can comprise varying levels of trust, depending on a published location of the application 126 and other factors. These varying levels of trust are described in greater detail as part of the discussion relating to the runtime 124, below.

A published location can be a location from which the application 126 is intended to be executed or cached. For example, if the builder 122 is building a requested trust level for an application that is to be available at a website, the published location for the application can be a URL indicating the domain from which the application can be accessed, such as that of the remote location 104. Similarly, if the builder 122 is building a requested trust level for an application that is to be accessed from a local source (such as the local memory 120), the published location for the application can be a filing system address from which the application can be accessed locally. Published locations can be used by the runtime 122 to aid it in determining appropriate trust levels at which to execute applications.

By building requested trust levels for applications, the builder 122 enhances security for computer systems. It also provides for a consistent user experience. Applications having requested trust levels can be executed at a consistent trust level regardless of where the application is executed from. By so doing, a user's experience can be consistent without regard to what computer or device from which the user executes the application. Also, applications having requested trust levels are more likely to behave robustly. These applications, because they are executed at a consistent trust level, are not subject to fluctuations due to being executed at a trust level at which they were not designed.

Executing an Application at an Appropriate Trust Level

Referring to FIG. 4, an exemplary process 400 for executing an application at an appropriate trust level is shown. The process 400 is illustrated as a series of blocks representing individual operations or acts performed by the tools 116 and/or the runtime 124. This process 400 can be implemented following the process 200 or can be implemented separately as a stand-alone process.

In the ongoing embodiment the application 126 is used for purposes of discussion. The application 126 can be received, accessed, executed, or cached from a remote source or locally, such as the remote location 104 and the local memory 120, respectively. The application 126 has a requested trust level, though that requested trust level may or may not have been built by the builder 122 as described in the process 200 above.

In some embodiments the process 400 begins when a user attempts to execute the application 126, such as by double-clicking on the application 126. When the user does so, the tools 116 can cache the application 126 to the cache 118 (shown) from an accessible location of the application 126.

Determining a Requested Trust Level

At block 402, a requested trust level for an application is determined. In the ongoing embodiment, the runtime 124 extracts from the application 126 an embedded requested trust level. The runtime 124 can determine whether or not the attributes 308, 310, or 312 are embedded in the application 126. Thus, if the runtime 124 determines that the attribute 308 of “requireFullTrust=yes” is embedded in the application 126, it determines that the application 126 requests the full trust level 302. If the runtime 124 determines that the attribute 310 of “trustLevel=Domain” is embedded in the application 126, it determines that the application 126 requests the location-dependent trust level 304. Similarly, if it determines that the attribute 312 of “trustLevel=Restricted” is embedded, the application 126 requests the restricted trust level 306.

Because the application 126 can contain malicious code, the runtime 124 does not trust the requested trust level of the application 126. For example, criminal persons might write applications having various requested trust levels using a copy of the builder 122, for instance. The requested trust level of the application 126, however, can be used by the runtime 124 to help determine an appropriate trust level for executing the application 126, if one exists.

Determining a Permitted Trust Level

At block 404, a permitted trust level for an application is determined. This permitted trust level can be independent of how an application is transmitted. Whether an application is received via email, or a floppy disk, or through other manners, the permitted trust level can be the same. Likewise, from where an application originates, such as from a website or another computer user, does not determine what trust level is permitted. Rather, a permitted trust level for an application can be determined based on from what location it is cached or executable and its published location.

In the ongoing embodiment, the runtime 124 determines the permitted trust level for the application 126. It can do so based on from what location the application 126 is executable or cached, a published location extracted from the application 126, and/or having a signed certificate. The runtime 124 can use the published location to aid in determining a permitted trust level, but the runtime 124 does not need to trust the published location or any other information extracted from the application 126, as will be apparent below.

Referring to FIG. 5, an exemplary table 500 setting forth exemplary permitted trust levels is shown. The trust levels shown are set forth as examples; other permitted levels can be used or defined. The exemplary trust levels comprise the full trust level 302, the location-dependent trust level 304, and the restricted trust level 306. The location-dependent trust level 304 can be further delineated, in this embodiment into three sublevels: a local machine trust level 502; an intranet trust level 504; and an Internet trust level 506. The local machine trust level 502 is a higher trust level than the intranet trust level 504, which is higher than the Internet trust level 506. Various potential execution locations for the application 126 are set forth at numeral 508. Whether or not the execution locations (“ELs”) 508 for the application 126 matches the published location (here shown with the attribute “LocationID=”) is shown at a column 510 of FIG. 5.

At block 404a, the location from which an application is executable or cached is determined. In the ongoing embodiment, the runtime 124 determines the execution location 508 for the application 126.

At block 404b, a published location for an application is determined. In the ongoing embodiment, the runtime 124 determines a published location for the application 126 by extracting this information from the application 126, if the application 126 contains a published location. Here the published location can be indicated with an XML attribute, such as “LocationID=Domain”, where “Domain” is a URL.

At block 404c, whether or not an application is installed or highly trusted is determined. In the ongoing embodiment, the runtime 124 determines whether or not the application 126 is installed or highly trusted. If it is, the runtime 124 follows the “Yes” path and permits local machine trust 502 or full trust 302, at block 404d. If not, it follows the “No” path to block 404e.

At block 404d, if the runtime 124 determines that the application 126 is installed and requests full trust, such as by extracting “requireFullTrust=yes”, shown in table 500 at 512, the runtime 124 permits the application 126 to be executed at full trust 302, shown in table 500 at 514. If the runtime 124 determines that the application 126 is installed but does not request full trust, such as by extracting “requireFullTrust=no”, shown in table 500 at 516, the runtime 124 permits (but not requires) the application 126 to be executed at local machine trust 502, shown at 518.

Also at block 404d, if the runtime 126 has determined that the application 126 is highly trusted, such as by being signed with a certificate (shown at numeral 520), the runtime 124 permits full trust 302, shown at 522.

At block 404e, an execution location (“EL”) and published location are compared. If the execution location and the published location match, the runtime 124 proceeds along the “Yes” path to block 404f. If not, it proceeds along the “No” path to block 404g.

At block 404f, location-dependent trust level 304 is permitted. In the ongoing embodiment, the runtime 124 permits either the machine level trust 502, the intranet level trust 504, or the Internet level trust 506, based on either the published location or the execution location. As set forth in FIG. 5, these location-dependent trust levels 502, 504, and 506 are permitted.

Assume, for example, that a user receives an email with the application 126 attached. Also assume that the user saves the application 126 to his or her local memory 120. At some later point, if the user attempts to execute the application 126, the runtime 124 will follow the process 400 to determine an appropriate trust level at which to execute the application 126, if one exists. In this example, assume that the runtime 124 determines, at block 402, that the application 126 requests location-dependent trust level 304 for a website on the Internet (e.g., the Internet level trust 506). The runtime 124 does not need the requested trust level to determine a permitted trust level, as the requested trust level is not trusted.

At block 404a, the runtime 124 determines that the execution location for the attached application 126 is the local machine memory 120. At block 404b, assume that the runtime 124 extracts the published, remote location for the website from the attached application 126. At block 404e, the runtime 124 determines that the execution location and the published location are not the same. Because of this, the runtime 124 permits only restricted trust level 306 (shown at 528). This ensures that the application 126 is not given too high a trust. In this example, the attached application 126 can contain malicious code; the attached application 126 could be built to request a trust based on a website and have a published location matching that website without either these being trustworthy. Because the application 126 did not necessarily originate at the website that it claims to have originated from, it is not trusted. Thus, the runtime 124 will not permit location-dependent trust 304 or full trust 302 (assuming the application 126 isn't highly trusted for some other reason).

In some cases, though, the runtime 126 permits location-dependent trust level 304. If an application is cached from the same location as published for the application, for instance, the runtime 126 will consider the application more trustworthy. If, for example, a user attempts to execute from a website the application 126, the runtime 124 can determine that the execution location of the application 126 is the website. If the application 126 also has a published location of this website (extracted by the runtime 124), which matches the execution location, the runtime 124 permits the application 126 to be executed at the Internet trust level 506. This is permitted because a trust level associated with that website is logical to permit; as the application 126 has been determined to actually be from that website (it has an execution location matching a published location of that website). That website can have a particular trust level associated with it that is set by an administrator or based on various factors analyzed using an algorithm, or through other well-known manners.

Determining and Executing at an Appropriate Trust Level

At block 406, an appropriate trust level is determined. The appropriate trust level can be determined based on comparing a requested trust level for an application with a permitted trust level. If an application has a requested trust level less than that of a permitted trust level, the runtime 124 can set the appropriate trust level as that of the requested trust level. If an application has a permitted trust level and requested trust level that are equal, the runtime 124 can set the appropriate trust level as that of the permitted trust level. If an application has a requested trust level higher than that of its permitted trust level, the runtime 124 can fail to set any trust level as appropriate. By failing to permit execution of an application at a lower trust level that it requests, the runtime 124 can limit inconsistent or non-robust operation of the application.

At block 408, an application is executed at an appropriate trust level, if one exists.

In the ongoing embodiment, the runtime 124 determines appropriate trust levels, which can comprise: the full trust level 302; the location-dependent trust level 304; or the restricted trust level 306. If no appropriate trust level exists, the runtime 124 will not execute the application 126 at block 408.

Referring to FIG. 6, an exemplary table 600 setting forth exemplary appropriate trust levels 602 are shown. These appropriate trust levels 602 shown are set forth as examples; other appropriate levels can be used or defined. The exemplary appropriate trust levels 602 at which an application can be executed comprise the full trust level 302, the location-dependent trust level 304, and the restricted trust level 306. The location-dependent trust level 304 is shown with further delineation, here the local machine trust level 502, the intranet trust level 504, and the Internet trust level 506. The table 600 shows one way in which the runtime 124 can determine appropriate trust levels 602 based on permitted trust levels shown in column 604 and requested trust levels of full trust, location-dependent trust, and restricted trust, shown in columns 606, 608, and 610, respectively.

If the requested trust level of the application 126 is the restricted trust level 306, the runtime executes the application 126 at that level. If the permitted trust level is higher than the restricted trust level 306, the application 126 likely can be executed and operate fully at the restricted trust level 306. As shown in the table 600, if the permitted trust level shown in column 604 is the local machine trust level 502, for instance, the runtime executes the application 126 at the restricted trust level 306 if that is requested (shown at 612). Various ways in which the runtime 124 can execute applications at the restricted trust level 306 and embodiments of this level are set forth in greater detail below in a section entitled, “Exemplary Restricted Trust Level.”

If the requested trust level is higher than the permitted trust level, the runtime 124 may not execute the application 126. Executing an application at a lower trust level that it requests can sacrifice robust and consistent operation of the application. This also can diminish a user's experience in using the application. Executing the application at above the permitted level can be dangerous, and so is not done. As shown in the table 600, if the permitted trust level shown in column 604 is restricted but the requested trust level shown at column 608 is location-dependent, the runtime 124 can fail to execute the application 126 (shown at numeral 614).

If the requested trust level is equal to the permitted trust level, the runtime 124 executes the application at the permitted/requested trust level. Examples of this are shown at numerals 616, 618, and 620.

Thus, the runtime 124 executes applications at appropriate trust levels. Applications may not be executed at higher trust levels than those at which they can be trusted. They can be executed at lower trust levels if they can be robustly and fully operated at these lower trust levels, based on a lower, requested trust level. And they can be executed at a permitted trust level if they can be trusted at this level and need to be executed at this level for full operation.

Exemplary Restricted Trust Level

Referring to FIG. 7, an exemplary process 700 for executing an application at an exemplary restricted trust level is shown. This restricted trust level permits execution of applications while prohibiting those applications from performing operations capable of endanger a user's computer or information. The process 700 is illustrated as a series of blocks representing individual operations or acts performed by the tools 116 and/or the runtime 124. This process 700 can be implemented as part of the process 400 or can be implemented separately as a stand-alone process. The restricted trust level set forth in this process is one implementation of the restricted trust level 306 described above.

At block 702, potentially damaging operations in an application are determined. In an ongoing embodiment, the runtime 124 scans the application 126 for custom code and/or any feature that requires connections to any data source outside of the application's 126 boundaries. The runtime 124 can do so by finding all URLs (e.g., links and website domains) in the application 126. These URLs can indicate that the application 126 is capable of accessing information or locations outside of the application 126 itself.

At block 704, potentially damaging operations are neutralized. In the ongoing embodiment, the runtime 124 neutralizes URLs found in the application 126, so that no data source outside the application boundaries can be contacted.

At block 706, rights potentially exercised for an application are removed. In the ongoing embodiment, the runtime 124 assigns a random execution location and/or published location to the application 126. By so doing, a trust level above restricted that potentially could be allowed for the application 126 due to its execution location or published location is removed.

At block 708, all custom code of an application is made safe. In the ongoing embodiment, the runtime 124 forbids and/or makes inaccessible all data connections (except email submittal), ActiveX controls, custom code written using managed code, roles, workflow, and the like in the application 126. Script is allowed only if it interacts exclusively with the data within the application.

At block 710, outside calls attempted during execution are intercepted and/or prevented. In the ongoing embodiment, the runtime 124 executes the application 126 but intercept and prevents any outside calls by the application 126 (such as to a URL).

If, for example, the application 126 is a form template but is to be executed at this restricted trust level, the application 126 can create information but cannot access any information other than the information that it creates. In the case of a form template, the runtime 124 executes the template and permits it to create an electronic document, receive data keyed into the electronic document from a user, and the like. The runtime 124 does not permit, however, the template from accessing or sending information outside of the application's boundaries, such as from or to a user's memory (e.g., the memory 114), an intranet site, or an Internet site.

At block 712, if the application 126 is rendering a view the runtime 124 assigns a fictitious URL to the view. This fictitious URL can have a very low level of permission. The runtime 124, following this low level, can prevent calls to external resources that the application 126's view is attempting to reach. In one embodiment, the view comprises Hyper Text Machine Language (HTML). HTML is a language that is capable of referencing URLs in many different ways, such as to script, styles, pictures, and frames. In part for this reason, the runtime 124 can perform additional operations to further secure the view, set forth at blocks 714 and 716.

At block 714, the runtime 124 traps outside calls, such as those attempted by the view that are not prohibited at block 712. In one embodiment, the runtime 124 traps outside calls by mapping all URLs through one or more particular code paths. Thus, these calls must use these code paths. The runtime 124 can, however, block these code paths, thereby prohibiting these outside calls from being made using these URLs.

In some cases, however, the application's 126 view is capable of making an outside call through a URL with a redefined interpretation.

At block 716, the runtime 124 finds and neutralizes URLs with a redefined interpretation. When URLs are interpreted in new ways, it can be difficult to prevent outside calls that use them. To aid in preventing these outside calls, the runtime 124 can scan a rendered view as it is updated to find these URLs. As the view is updated, the runtime 124 determines whether or not URLs are being interpreted in a new way. If so, the runtime 124 neutralizes these URLs, such as by deleting them from the view.

In one embodiment, the view comprises HTML. In these cases, a URL can be interpreted in a new way with a “base tag”. The runtime 124 can delete base tags that redefine how a URL is interpreted from the HTML code of the view.

Conclusion

The above-described tool enables execution of applications at appropriate trust levels. Although the invention has been described in language specific to structural features and/or methodological acts, it is to be understood that the invention defined in the appended claims is not necessarily limited to the specific features or acts described. Rather, the specific features and acts are disclosed as exemplary forms of implementing the claimed invention.

1. A computer implemented method comprising: determining potentially dangerous operations of an application by finding one or more universal resource locators (URLs) in the application and analyzing code associated with each URL effective to determine whether the code is configured to communicate with one or more remote locations; and preventing or intercepting the potentially dangerous operations of the application by embedding a requested trust level into the application, the requested trust level comprising at least one of: a full trust level that requests permission to communicate with any remote location; or a location-dependent trust level that requests permission to communicate with at least one of the one or more remote locations.

2. The method of claim 1, wherein the act of preventing or intercepting comprises preventing any information from being sent outside of, or received by, the application.

3. The method of claim 2, wherein the application comprises arbitrary HTML.

4. The method of claim 1, further comprising assigning a fictitious universal resource locator (URL) to a view being rendered by the application.

5. The method of claim 4, wherein the fictitious URL is associated with a low level of permission.

6. The method of claim 1, further comprising trapping outside calls attempted by a view being rendered by the application.

7. The method of claim 1, further comprising finding and neutralizing a universal resource locator (URL) having a redefined interpretation that is in a view being rendered by the application.

8. The method of claim 7, wherein the view comprises Hyper Text Markup Language (HTML) and the act of finding and neutralizing the URL comprising finding and neutralizing a base tag.

9. The method of claim 1, wherein the act of preventing or intercepting comprises preventing or intercepting all outside calls made by the application, and wherein the outside calls comprise a call made by the application to a universal resource locator outside the application's boundaries.

10. The method of claim 1, further comprising neutralizing the potentially dangerous operations.

11. The method of claim 1, further comprising removing rights potentially exercisable by the application.

12. The method of claim 11, wherein the act of removing rights comprises assigning a random execution location or published location to the application.

13. The method of claim 1, further comprising making safe custom code of the application.

14. The method of claim 13, wherein the act of making safe comprises forbidding or making inaccessible data connections or custom controls.

15. One or more computer storage media having computer-executable instructions for performing the method recited in claim 1.

16. A computer implemented method comprising: determining potentially dangerous operations in an application, the determining including scanning the application for universal resource locators and code that is configured to communicate with one or more remote locations associated with the universal resource locators; and preventing or intercepting the potentially dangerous operations by embedding a requested trust level into the application, the requested trust level comprising at least one of: a full trust level that requests permission to communicate with any remote location; or a location dependent trust level that requests permission to communicate with at least one of the one or more remote locations.

17. The method of claim 16, further comprising removing rights potentially exercisable by the application.

18. The method of claim 17, wherein the act of removing rights comprises assigning a random execution location or published location to the application.

19. The method of claim 16, further comprising making safe custom code of the application.

20. The method of claim 19, wherein the act of making safe comprises forbidding or making inaccessible data connections or custom controls.

21. The method of claim 16, wherein the act of preventing or intercepting comprises preventing or intercepting all outside calls made by the application.

22. One or more computer storage media having computer-executable instructions for performing the method recited in claim 16.

23. A computer implemented method comprising: determining if an application is configured to access data outside of the applications boundaries based on one or more URL's; responsive to determining that the application is configured to access data outside of the application's boundaries based on said one or more URL's, embedding a location dependent trust level into the application that requests permission to access data outside of the application's boundaries; and responsive to determining that the application is not configured to access data outside of the application's boundaries based on said one or more URL's, embedding a restricted trust level into the application that does not request permission to access data outside of the application's boundaries.

24. The method of claim 23, further comprising determining the operations and neutralizing the operations.

25. The method of claim 23, further comprising removing rights potentially exercisable by the application.

26. One or more computer storage media having computer-executable instructions for performing the method recited in claim 23.