A Study on resource discovery mechanism in distributed information networks

在诸如数字图书馆等应用中，用户倾向于在不知道对象的实际位置的情况下访问和检索对象，例如科学论文、图片和数字数据的页面。本研究将这些资源抽象为信息资源，通过对特定信息资源属性的描述来寻找资源的标识，在实际的互联网环境中，用户不得不使用URL或其他位置信息，因为它们是信息资源的标识，显示了存储位置。然而，要获得合适的识别符并不容易。提出了统一资源规范的新概念，提供了持久的解析由属性规范…组成的对象描述的能力本研究所采用的软件架构基于客户端-服务器模型，并采用了ANSI Z39.50协议标准。我们需要的是一个用于实验目的的轻量级协议，因此，我们修改和简化了Z39.50用于名称解析目的，名称解析服务器根据客户端提供URL的请求返回信息对象的标识符。该解析采用的机制是基于规则的算法的简单方法。从实际应用的角度来看，还需要兼容基于HTTP的访问，因此还实验实现了HTTP服务器和URS解析器之间的网关机制。从1996年到1997年，我们一直在进行研究，我们认为我们的原型实现表明了所提出的概念的可行性。对原型的细节的评估仍在进行中，以便进一步发展该方法。较少

The aim of this research was to investigate methods for managing, identifying and locating information which is distributed over a wide-area network, such as the Internet.In applications such as digital libraries, users tend to access and retrieve objects, e.g., pages of scientific paper, pictures, and numerical data without knowing actual locations of the objects. In this research, these materials are abstractly termed as 'information resource' and a general approach to look for identifiers of resources based on the description of attributes of the specific information resources.In actual environment in the Internet, users have to use URLs or other location information, since these are identifiers of information resources which shows the storage locations. However, it is not easy to get appropriate identifiers. The new concept of URS (Uniform Resource Specification) is proposed and this gives a persistent capability to resolve object description composed of the attribute specification … More into URL,thus allowing users to issue a query without knowing exact location of information.The software architecture employed in this investigation is based on a client-server model, and we adopted ANSI Z39.50 protocol standard for this purpose. What we needed is a light-weight protocol for experimental purpose, therefore, we modified and simplified Z39.50 for a name resolution purpose.The name resolution server returns an identifier of information object, upon the request from a client giving URS.The mechanism employed for this resolution is a simple method according to rule-based algorithm. From a practical aspect, compatibility with HTTP based accesses is also required, thus a gateway mechanism between a HTTP server and a URS resolver was also experimentally implemented. conventional HTML browser can be used to access to URS resolver through this gatewasy.The research was pursued through 1996 to 1997 and we think that our prototype implementation shows the feasibility of the proposed concept. Evaluation of the details of the prototype is still going on for further evolution of the approach. Less