Google Scholar Blog: 2011

Google Scholar Citations Open To All

Wednesday, November 16, 2011

a limited release of Google Scholar CitationsClick here and follow the instructions to get startedresearchers at US universitiesresearchers interested in genomicsAlex VerstakAnurag Acharya[alex verstak]Posted by: James Connor, Software Engineer

Shall I compare thee to... a cytotoxicity assay?

Monday, August 15, 2011

To sleep, perchance to dream (Hamlet)Nocturnal eating syndromes: to sleep, perchance to eatTo sleep, perchance to leakMore...To be, or not to be, that is the question (Hamlet)To slow or not to slow: the economics of the greenhouse effectTo normalize or not to normalize? Fat is the questionMore... Much Ado About NothingThe placebo: Is it much ado about nothingProgrammed ribosomal frameshifting: Much ado about knotting!More...Romeo, Romeo, wherefore art thou, Romeo (Romeo and Juliet)Wherefore art thou r3579x?: anonymized social networks, hidden patterns, and structural steganographyWherefore Art Thou Copper? Structures and Reaction Mechanisms of Organocuprate Clusters in Organic ChemistryMore...Et tu, Brute? (Julius Caesar)Et Tu, Fair Use-The Triumph of Natural-Law CopyrightEt tu, acetaminophen?More...All that glisters is not gold (The Merchant of Venice)All that palsies is not Bell's—The need to define Bell's palsy as an adverse event following immunizationAll that wheezes is not asthmaMore...Posted by: Darcy Dapra, Partner Manager

Google Scholar usage by browser

Thursday, August 4, 2011

As webmasters dabble into the land of exciting possibilities enabled by recent developments in browser technology, it's essential to remember that different users use different browsers, and webpages that work in one browser will not necessarily work in another. But developer time is limited - so, how important is it for a scholarly website to support version N of browser B? To help answer this qestion, here's the breakdown of Google Scholar searches by browser, fresh from the logs.Fraction of Google Scholar searches by browser for periods ending August 2, 2011.

year

month

day

Internet Explorer

59%

47%

46%

Firefox

25%

23%

Chrome & Safari

10%

19%

20%

Unknown

10%

Other

And now for the fine print...

This table doesn't necessarily represent trends in browser adoption, because Google Scholar usage is qualitatively different throughout the year. In particular, summer numbers do not include frantic searches by thundering herds of students studying for their exams; but annual numbers do.

These numbers are certainly heavily influenced by English searches. However, over half of our usage is from outside the US, and non-English speakers tend to search in their own language. Yes, for papers too; yes, I checked; and yes, some of their searches are in English. All things considered, I think the world stats should work well for most websites.

As of last month, versions 7, 8 and 9 of Internet Explorer were used in 18%, 21% and 3% of all searches, respectively. Versions 5 and 6 carry some, uhh, surprises for developers, but luckily their usage has dropped from 3% + 15% = 18% in the annual stats to 0% + 5% = 5% in the monthly stats. So, that settles it - we will only test Google Scholar with IE7+ going forward.

Firefox-wise, 10% of total usage is from version 3, and the vast majority of the remaining 13% is from versions 4 and 5. All three versions are quite similar in their rendering of HTML, so it's often sufficient to cover just the latest version in testing.

I have combined the two Webkit-based browsers, Chrome and Safari, because their rendering of HTML is nearly identical. For minor changes, we often test the latest version of Chrome and skip Safari, because Chrome's usage is currently 15% vs. Safari's 5%. Both of these browsers tend to be reasonably up to date - the vast majority of their searches last month came from Chrome versions 12 and 13 and Safari version 5.

All mobile browsers combined currently make up a very small fraction of searches, though whether one can call millions of searches "very small" is certainly a matter of perspective. The same applies to Opera and to a number of other browsers and platforms. Occasional testing of infrequently used browsers is key to a reasonable balance between serving your users and taking forever to rollout.

Bottom line? We recommend that you test your scholarly website with at least three browsers - Internet Explorer 7+, Firefox 3+, and Chrome 12+. If it doesn't work with either one of these browsers, then it likely doesn't work for a lot of your users.Posted by: Alex Verstak, Software Engineer

Google Scholar Citations

Wednesday, July 20, 2011

h-indexi-10 indexAlex VerstakAnurag AcharyaRichard FeynmanPaul DirachereAlbert EinsteinMargaret MeadAlonzo Churchlearn more about Google Scholar CitationsUpdate:sign upPosted by: James Connor, Software Engineer

Work Locally, Search Globally

Wednesday, July 6, 2011

Posted by Mohit Rajani, Software Engineer

A scholarly discovery, just in time for Father’s Day

Monday, June 6, 2011

Acute Histoplasma PericarditisPosted by Darcy Dapra, Partner Manager

Better highlighting for legal cases

Tuesday, February 8, 2011

We used to have colorful highlighting for cases. Some users have told us that this is hard to read. We now highlight legal cases more gently and provide an option to remove highlighting entirely. Happy reading!

Posted by James Connor, Software Engineer

Search opinions from specific courts

Tuesday, January 11, 2011

in the Southern District of New Yorkin the Northern District of California.advanced search page.oysters in tax courtshealth insurance in Fourth Circuit courts.recent opinions about oil spills in Louisiana state and federal courtscases about election recounts in Florida, Minnesota and Alaska courts.Posted by Chris Hundt and Itai Gurari, Software Engineers

Scholar Blog