Hook said: “Dimensions is not intended to be a bibliometric product, but rather a discovery product”. (The Bookseller, 15th Jan 2018)
It’s about discovery and it’s public, so I gave it a quick test:
mongolian folk song (choosing the “full data” option).
Getting Dimensions to “Limit to” Open Access and reload results according, was a fiddly multi-click operation. But when this was done, none of the results were relevant. It even managed to surface an article touching on Du Bois (author of the 1903 book Souls of Black Folk), a highly spurious result known to me from previous group tests, and which appears to arise from some antiquated synonym algorithm that states “Mongolian = black”. The best Dimensions could do, in over 100 results, was the tangentially relevant paper “Translocal English in the linguascape of Mongolian popular music”, on the use of the English language in the music scene of post-socialist Mongolia. This article’s record page gave a valid link which eventually got me to the full article at Wiley.
I then removed their OA filter, which actually improved OA results! However the top results were then questionable. The top two results were relevant, but both were from the Canadian Center of Science and Education. The third result was also relevant, but from the Atlantis Press. Despite being OA, none of the sources for these three results are indexed by the DOAJ or JURN. The Scholarly Kitchen suggests that Dimensions users may be finding more such results in their search…
“Dimensions is inclusive in terms of content coverage, rather than curated as is the case for Scopus and Web of Science. Of course, what reads to some as more inclusive can be seen by others as less rigorous selection…”
On both search types the results rapidly devolved into off-topic medical and a few natural-history items, although there was one relevant book review lurking one set of ‘the first 100’ results, and a possibly-relevant chapter. Sadly the review article was found to be on the ‘academia-only’ service Project Muse, and as such all but the first page was blocked to the public. The book chapter on “Tourism and culture in Mongolia” was perhaps at Elsevier, but on arrival Elsevier’s link gave the response that “Your request cannot be processed at this time”, presumably due to my lack of log-on credentials.
If Dimensions had been in my recent group-test based on these keywords, it would have been nowhere in the rankings, with a score of “1” for “Translocal English in the linguascape of Mongolian popular music”. And even then it’s a marginal result, and so I would have been being charitable to Dimensions.