“Search Needs A Shake Up”

“Search Needs A Shake Up” writes Oren Etzion in an op-ad article in the comments column of Nature (4th Aug 2011). The under-the-hood discovery technology has not significantly progressed, even while more-or-less helpful widgets have proliferated on search pages, and search results are transformed on-the-fly to fit almost any mobile device. Most ordinary people still want their search-engine to be like a ‘magic oracle’ — ask a natural language question, and get a one-line correct answer back. Of course, that doesn’t work — which is why Yahoo Answers and its more professional imitators are so popular for those with few search skills. The problem is, the laziness of their users makes unpaid slaves of their helpers.

But until we get that magic one question / one answer solution (Etzion outlines some research on that, but don’t hold your breath) how about just teaching people to search properly, ideally intensively and from primary school level onwards? It’s not rocket science. It’s no more difficult than learning the basics of the Highway Code by heart, or some basic smatterings of Spanish. Doing quality search should be as natural as basic literacy. Once the school-level training is bedded down, then refuse entry to university applicants who cannot pass a rigorous one-hour “search and find” test. The kids and their teachers will soon get the message.

Firefox to Chrome

I finally made the switch from the Firefox web browser to Google Chrome. My detailed instructional is here, which includes how to disable most of the “+1 button” gunk on Google Search results pages.

OKF proposes new search-engine

The Open Knowledge Foundation adds a “Website for finding and searching OA humanities journals” to its Ideas Incubator.

Sidebar checked

The JURN blog’s sidebar has been checked for dead or lapsed blogs, run-out-of-funding projects etc. Cleaned, and added some new links.

Ion ISC02 Book Saver goes to Amazon pre-order

Amazon.com are now taking U.S. pre-orders for the Ion ISC02 Document Scanner. No sign of it on Amazon U.K. yet. ETA is rumoured to be December 2011. At $190 it’s going to be a bit more expensive than the $150/£129 it was first touted as back in January 2011. I’d guess about £159 in the UK, if Amazon UK ever gets them in stock.

I’d imagine that images captured at that distance won’t have the massive resolution that standard OCR software is used to, unless Ion have something special happening with the camera lens (an array of multiple cheap lenses for gigapixel capture?). But if general software such as Adobe Acrobat Pro can OCR even tiny footnote text from the page images, then the Ion ISC02 is going to be a winner.

DOAJ article-level indexing leaps over 3,000 journals

The Directory of Open Access Journals (DOAJ) team leaps over the “3,000 journals indexed at article level” line. The press release was actually for 7,000 in the Directory listings, but only 45% of these are indexed/searchable at article level. Hence the DOAJ is now at about 3,100 in terms of being a search tool, and must have crossed the 3,000 line sometime in the late summer.

At current count, 905 of the Directory records are for journals in the arts and humanities (1081 if you include linguistics), with an unknown number of that subset searchable at article level. If the “45% indexed” figure holds true for arts and humanities, the DOAJ might be searching around 450 such journals at the article level.

The Location of Academic Knowledge

The Location of Academic Knowledge (journals by country) included in Web of Knowledge ^*. Click for larger version.

From: Graham, M., Hale, S. A., and Stephens, M. (2011) “Geographies of the World’s Knowledge”. Oxford Internet Institute & Convoco.

  * Web of Knowledge subsumes Web of Science, and includes Journal Citation Reports, Science Citation Index (1945+), Social Science Citation Index (1956+) and the Arts and Humanities Citation Index (1994+). ‘High impact’ journals, inc. around 1,300 major arts and humanities titles in Arts and Humanities Citation Index. One might keep in mind, however, the 2009 Gale Reference Review comment that… “I looked at the widely touted figures [for journal numbers] in the promotional materials [ of Web of Science and Scopus and found ] they should not be taken for granted. Many of these are incorrect and exaggerated. Their compilation has been fast and loose, sometimes making them fiction rather than fact.”

JURN checked and updated

I’ve checked the JURN Directory using Linkbot:—

* Discovered and fixed about 15 broken URLs, and also fixed these in the Search URL database if needed.

* Added 18 new journal titles.

* Removed from the Directory and Search URLs database, because dead: Thlazine; Cybermetrics; Altitude; Inbhear; Broadsheet; Yemen Update; Mouseion (gone to live behind the Project Muse wall, seemingly?); Italian Politics and Society; Journal of Philosophical Practice; Views; Text, Practice and Performance; Digital Document Quarterly.

JSTOR free before 1923

JSTOR are offering free worldwide access to journal articles published in the USA before 1923. This is 6% of JSTOR’s content. No log-in required, no Hathi-like fuss about “which country are you in, because your government might have passed a ridiculously extended copyright law”. Just go to Advanced Search, tick the box for “Include only content I can access”…