Several people have asked me implicitly whether the Open Mirror is a portal. My first response, which is true but not substantive, is that I don’t yet know what the Open Mirror is; we are conducting a feasibility study to identify what, if anything, it should be. But the “portal” question is interesting. If all the UK’s open access research material is collected together, with some reasonable metadata, then the obvious first thing to do is to put a search box on top of that and, of course, that search could also cover UK open access research that, for whatever reason, was not brought under a single roof. There are some good reasons why a search box, such as Google’s search service, might want a local cache of the content it is searching: to build an efficient index; to produce snippets and thumbnails; to enhance the search in various other ways. A search box might be the obvious thing to do, but is it the right thing?
To return to an age-old question, has Google solved the internet discovery challenge? The “discovery” argument in favour of the Open Mirror is that there is room for complementary services, and that a national aggregation of open access material can contribute to those services. This might be by offering an API, by close working with library web-scale discovery engines, such as Summon, by surfacing the UK’s open access research in the best way possible to attract the attention (and indexing) of major search engines, or by joining with similar initiatives elsewhere (Australia, Netherlands, the US) in some ways. It seems inconceivable that researchers would want to search a UK-specific corpus of material but, if that UK corpus were working behind the scenes to make UK research more visible to all potential readers, then that might be a valuable service. If that’s the right approach, then does the Open Mirror itself need credibility with the academic community? Does it need its own search box, or even its own web presence?
Of course, the Open Mirror could have other value than to improve discovery. It might contribute to preservation, or management information, or easier text-mining and exploitation of open access research. Those will be the topics of future posts here.