{"title":"Second-best practices for interoperability","authors":"Martin C. Libicki","doi":"10.1145/230871.230877","DOIUrl":null,"url":null,"abstract":"■ If the GII is to realize its full potential, it will have to support distributed applications among heterogeneous user systems exchanging not only bits, but also mutually understood meanings (nouns and verbs). This requires standards. Yet the best method of generating such standards, through explicit consensus, may simply not work well, or on time. Second-best approaches such as middleware, lexical primitives, or metalanguage need to be considered. he vast world of information technology standards may be characterized by its two largest realms: public communications and private computation. Public communications entails hauling bits (with the requisite level of service, block definition, reliability and security) among entities that may be anonymous to each other. This is the province of the telephone system, the Internet, etc. Private computation is epitomized by the fully functional corporate network maintained by a specified office to support applications using specific and well-understood information definitions. Both realms are characterized by specific standards. Interoperability in public communications is supported by ITU and Internet standards. Portability of applications across heterogeneous architectures is supported by an ad hoc mixture of language standards, data-item standards, operating system standards, and emerging application portability interfaces. Communications tends to get the standards it needs—which it must if public communications is to exist. Computer uses tend to be covered by standards less often, but privately managed computer systems can use hand-crafting and tight management to get over the bumps. Over the next ten years, the formation of a global information infrastructure—the great challenge in information technology—will require a merger of the two realms. That is, public systems will need to find ways of exchanging and interpreting not only bits, but meanings. They will have to find ways of referring to common concepts using, if not identical vocabulary, at least a vocabulary that permits translation. Moreover, they will have to exchange information without the labor-intensive pre-negotiation usually entailed in the construction of private infrastructures. One can already glimpse applications that run over heterogeneous equipments (that is, nodes) operated by heterogeneous owners to transfer and understand information. Network management is an early application, albeit one well-standardized by its origin in telecommunications. More typical in the future may be environmental monitoring. An adequate environmental picture may require the fusing of data from ground sensors, water sensors, airborne laser-fed Second-Best Practices for Interoperability","PeriodicalId":270594,"journal":{"name":"ACM Stand.","volume":"268 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1996-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"ACM Stand.","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/230871.230877","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
■ If the GII is to realize its full potential, it will have to support distributed applications among heterogeneous user systems exchanging not only bits, but also mutually understood meanings (nouns and verbs). This requires standards. Yet the best method of generating such standards, through explicit consensus, may simply not work well, or on time. Second-best approaches such as middleware, lexical primitives, or metalanguage need to be considered. he vast world of information technology standards may be characterized by its two largest realms: public communications and private computation. Public communications entails hauling bits (with the requisite level of service, block definition, reliability and security) among entities that may be anonymous to each other. This is the province of the telephone system, the Internet, etc. Private computation is epitomized by the fully functional corporate network maintained by a specified office to support applications using specific and well-understood information definitions. Both realms are characterized by specific standards. Interoperability in public communications is supported by ITU and Internet standards. Portability of applications across heterogeneous architectures is supported by an ad hoc mixture of language standards, data-item standards, operating system standards, and emerging application portability interfaces. Communications tends to get the standards it needs—which it must if public communications is to exist. Computer uses tend to be covered by standards less often, but privately managed computer systems can use hand-crafting and tight management to get over the bumps. Over the next ten years, the formation of a global information infrastructure—the great challenge in information technology—will require a merger of the two realms. That is, public systems will need to find ways of exchanging and interpreting not only bits, but meanings. They will have to find ways of referring to common concepts using, if not identical vocabulary, at least a vocabulary that permits translation. Moreover, they will have to exchange information without the labor-intensive pre-negotiation usually entailed in the construction of private infrastructures. One can already glimpse applications that run over heterogeneous equipments (that is, nodes) operated by heterogeneous owners to transfer and understand information. Network management is an early application, albeit one well-standardized by its origin in telecommunications. More typical in the future may be environmental monitoring. An adequate environmental picture may require the fusing of data from ground sensors, water sensors, airborne laser-fed Second-Best Practices for Interoperability