Abstract: A redirection object may be obtained, based on a current request of a current requestor. The redirection object request may be associated with a response to a previous request for a redirection to a destination network location, the previous request associated with a previous requestor. A current transmission origination location associated with a current requestor transmission of the current request may be determined. It may be determined whether the current transmission origination location includes a similarity to a previous transmission origination location associated with a previous transmission of the previous request, based on identification information included in the redirection object. It may be determined whether to initiate a redirect operation of the current requestor to the destination network location, based on a result of the determining of the similarity between the current transmission origination location and the previous transmission origination location.

Abstract: One or more techniques and/or systems are provided for selective server-side tracing of user requests that match tracing criteria. For example, internal server processing data and/or a server-side perspective of web traffic data associated with a server processing a user request may be generated server-side as tracing data. The tracing data may be utilized in troubleshooting issues that may arise between a client machine and the server (e.g., error in uploading a photo to a website hosted by the server). Because a web-based resource (e.g., website) may be hosted by a large number of servers configured to process numerous user requests over time, an overwhelming amount of tracing data may be produced in the absence of weeding out requests for which tracing data is not needed. Thus, utilizing tracing criteria to selectively trace particular user requests promotes efficient troubleshooting.

Abstract: DNS-based content routing techniques are described. In one or more implementations, data is examined that describes interactions via a network with content via a domain name. Responsive to the examination, a policy is adjusted to change how one or more network addresses are resolved for the domain name for access to the content. A communication is formed that includes the adjusted policy to be communicated to one or more domain name system (DNS) servers, the adjusted policy configured to specify which network address are resolved for the domain name by the one or more DNS servers for access to the content.

Abstract: Activities of users of a service often involve one or more resources, such as uploading or downloading files in a file system of an FTP server. The activities of the users may be tracked and recorded in an activity log in order to identify frequently performed activities involving particular resources, and for such frequently performed activities, one or more activity actions may be performed. For example, malicious users may upload or utilize an equivalent set of assets stored in several accounts. The frequency of these undesirable activities may be identified, and an activity action may be automatically applied to the users (e.g., banning accounts), resources (e.g., deleting assets), and/or activities (e.g., blocking access to the resources). Conversely, desirable activities involving particular resources may be similarly detected, and the activity action applied to such desirable activities may involve reporting the desirable activity to an administrator of the service.

Abstract: The creation and utilization of network restore points is provided by a schema associated with a structured data set that can be conveniently backed up by periodically taking snapshots of the structured data to establish a series of restore points that can be used in the event that the primary structured data becomes lost or corrupted. In general, the snapshots are only taken after the structured data set has undergone a change in content, although they may be taken at other times as well. The snapshot may be taken by a network restore point engine that can be accessed by a user over the Internet. In addition, the snapshot may be stored on an Internet-based storage medium or it may be pushed to or pulled by the client so that the snapshot resides on a client-resident storage medium.

Abstract: The creation and utilization of network restore points is provided by a schema associated with a structured data set that can be conveniently backed up by periodically taking snapshots of the structured data to establish a series of restore points that can be used in the event that the primary structured data becomes lost or corrupted. In general, the snapshots are only taken after the structured data set has undergone a change in content, although they may be taken at other times as well. The snapshot may be taken by a network restore point engine that can be accessed by a user over the Internet. In addition, the snapshot may be stored on an Internet-based storage medium or it may be pushed to or pulled by the client so that the snapshot resides on a client-resident storage medium.

Abstract: Various embodiments provide an evaluation module that is configured to monitor activities of new users and ascertain, from monitored activities, a reputation associated with the new users. In at least some embodiments, the evaluation module comprises an activity store and/or a reputation service. The activity store can monitor different types of actions associated with new users. For example, the activity store can monitor content-based actions that pertain to particular pieces of content and/or user behaviors associated with new users. The reputation service, which may or may not comprise part of the evaluation module, can ascertain from information provided from the activity store, information, such as a reputation category, associated with content-based actions and/or user behaviors associated with new users. Based on a new user's information, e.g. reputation category, the new user can be assigned an activity type which, in turn, defines allowed and disallowed activities for the new user.

Abstract: Techniques and environments that increase the convenience, efficiency and variety of cloud services are offered to clients. User data having an original representational format is losslessly transformed to form one or more alternate renditions having various representational formats based on the lossless transforms. The renditions, which may be pre-generated, can be stored on a network server instead of the identical user data to thereby reduce redundant computation and storage costs. In some cases both the renditions and the original use data may be stored. The original user data may be reconstructed, recreated, or restored using the alternate renditions.

Abstract: The creation and utilization of network restore points is provided by a schema associated with a structured data set that can be conveniently backed up by periodically taking snapshots of the structured data to establish a series of restore points that can be used in the event that the primary structured data becomes lost or corrupted. In general, the snapshots are only taken after the structured data set has undergone a change in content, although they may be taken at other times as well. The snapshot may be taken by a network restore point engine that can be accessed by a user over the Internet. In addition, the snapshot may be stored on an Internet-based storage medium or it may be pushed to or pulled by the client so that the snapshot resides on a client-resident storage medium.

Abstract: An arrangement for reducing the occurrence of harmful cross-site scripting is provided by segregating on-line content or other resources so that they are accessible at different domains or subdomains, each of which corresponds to a set of users, called a “sharing set,” where each user in the set has identical access privileges to certain resources. The sharing set is provided with an identifier (which may or may not be unique), so that the identifier may be used as the name of the domain or subdomain for which any member of the sharing set is authorized to access the resources located there. In this way, script that is embedded with the content can only be executed among members of the sharing set. Users who are not members of the sharing set are unable to invoke cross site-scripting attacks that would allow them to gain access to data from sharing set members.