In the case of the Internet, there was first the major effort of the US Defense Advanced Research Projects Agency (DARPA), which invested twenty years of continuous development (first with NCP then TCP/IP). It has had a slower uptake in part because there has been no “major force” backing it. But in a broader use of the word, articles, software, workflows, etc., are all data or DOs.Īs mentioned above, semantic software in the context of the Web and elsewhere is twenty years old. As stated above, “open” scientific data are part of OS along with articles, software, workflows, etc. The conference created the acronym FAIR data, identifying some of the characteristics that OD need to be maximally useful: findable, accessible, interoperable, and reusable. Several years after that, a conference at the Lorentz Center in Leiden University asserted that merely “opening” that data was not nearly enough to assure interoperability. Several years later, the US Office of Science and Technology Policy tasked the research-oriented agencies to make all scientific articles and data created with Federal support available to the public (Open Data (OD)) for the cost of an Internet connection. The thesis of this group was that we could now store more data than we could effectively process. At that time, the US Federal agencies established an interagency senior sterling group to research big data. As stated above, the hardware innovation trigger has been in evidence for ten years. Gartner asserts that this curve represents a common pattern that arises with many new technologies or other innovations. In such a technical environment, OS will be enabled. As the word “data” is used here, it includes articles, data, software, workflows, etc. When efforts such as these or others achieve their goals, we will have the third generation of computing, where not only is the network the computer, but, as indicated above, “all” of its data are interoperable. It was also designed several decades ago, but it, too, has only become practical with today's high performance connected computers. Another effort making progress in the interoperability of heterogeneous data is the Digital Object (DO) Architecture. Twenty years ago when Web designers created the Semantic Web, an observer noted that “the only new thing about the Semantic Web is the Web.” Nevertheless, just as the Web revolutionized access to information, the Semantic Web may revolutionize access to knowledge. Semantics is one of the long-studied areas of artificial intelligence and it entered a new phase in the new century. Interoperability of data requires a semantic layer. Publication takes on a new meaning here, since the computers must “understand” all of these artifacts well enough to provide us novel assistance. Hardware with these capabilities is the innovation trigger that will enable Open Science, which is the “publication” of all relevant science results: data, software, workflows, etc., in addition to summary articles describing those science results, and accessibility of all these elements (at least all those produced with public money) for the cost of an Internet connection. They are a thousand times more powerful and a thousand times cheaper than computers were a generation ago: A multi-terabyte disk costs a hundred dollars, a thousand dollar computer can execute several billion instructions per second and the network that connects millions of such computers has a bandwidth approaching a terra-bit per second. High performance networked computers are today's equivalent of the printing press. The printing press was the innovation trigger that enabled semi-Open Science (OS).
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |