Category Archives: internetlibrarian

CrowdAsk

A Gamified Crowdsourcing Reference System for Libraries Developed by Purdue Libraries. Ilana Stonebraker, Tao Zhang

  • FAQ system where users can vote and contribute.
  • Points, badges, and leaderboards
  • 122 questions, 232 answers since spring 2014
  • Open source and openly available
  • Helping users learn together and from each other
  • Website

Had a FAQ help system that wasn’t working well. How to know which users want what in help? Started a research project and surveyed students to see what they’re looking for in help.

In traditional library reference, all questions are treated equal. Majority of reference questions are lower level. Reference service model is flawed; students getting answers from librarians, professors, other students, and friends. We only know about the answers from the librarians, not from the other sources.

Solutions in crowd sourcing is the utilization of content experts, like graduate students. Novices learning together better reflect a participatory culture, metaliteracy. Benefit is that it’s a single channel, and focus is on librarians as community builders versus information sources.

Types of questions asked were course-related, CrowdAsk related, library services or resources, how-to, and conceptual. Motivated by reciprocity, not by the points.  It’s about cultivating community, rather than gamification.

Offered as a secondary choice, after the Ask the Librarian choice on the website. Staff ensures that questions are answered in one day. Goal: develop sustainable user engagement and community involvement as part of the Purdue University Libraries website.

Looking for partners and test cases; GNU General Public License Version 2 on GitHub.

Advertisements

Leave a comment

Filed under internetlibrarian

Search: social, personal, & everywhere

Greg Notess, Montana State University @notess

Personalization

  • Pros:
    • customize results ranking
    • increased relevance
    • targeted ads
  • Cons
    • librarian searching
    • privacy
    • not always relevant

The Filter Bubble: concerns based on results personalization. People will only see material that they agree with; if searching for liberal (or conservative) information that match that point of view.

You get personalized results whether you’re logged in or not – Google will remember for you. Will see previous searches, GeoLocation, YouTube viewing history, all pages visited (if toolbar enabled.)

Google behavioral ad targeting: google.com/settings/ads. Logged in or not. Check and see how accurate it is on targeting you. You can disable it by logging out. Easiest way is to use the private or incognito mode. Even in private mode, there is some personalization, like GeoLocation.

Rapleaf: information based on your email address. BlueCava: device tracking.

Alternative search engines: duckduckgo.com, dontbubble.us, ixquick.com

Social searching: Facebook – private to a point. Makes it harder to search. Twitter – can have private tweets. LinkedIn – default shares who searches for you; alternative search approaches: cached search options, Bing search results.

Facebook graph search: combine odd combinations of factors (republicans who love sailing and live in Minneapolis.) Can narrow search by sex, age, employer, etc. Dependent on what information people share.

Twitter searching: follow hashtags. Topsy: will search full Twitter feed. Hshtags.com – app to search hashtags. Twitter: #{name} filter:links.

Unshorten – unshorten the link, to determine what it is. wheredoesthislinkgo.com, knowurl.com, clybs.com/urlexpander

Map/Geographic search: Google and Bing maps both have street view, different data sources, different imagery dates. Google Earth has the ability to browse back in time.

World.time.com/timelapse – time-lapse imagery. Earthengine.google.com – environmental data and analysis.

Leave a comment

Filed under internetlibrarian

Super Searcher Tools & Tips

Mary Ellen Bates, Principal, Bates Information Services, Inc.

@mebs

Google dictionary now has word origin, translate, and usage over time.

Google autocomplete can be used to find alternatives to a product or service. Ex: evernote vs. – and then wait, and Google will list alternatives.

snipr.com/gdork14 – links to Google dorking.

Test your org’s vulnerabilities: filetype:csv OR filetype:xls OR filetype:log site:your.org intext:username OR email OR confidential OR…

Change order of search words: ocean fishing acidification VS. acidification fishing ocean. Got different number of search results with the same words in the query, just a different order. Used Bing, and got different results. Each query resulted in some results that didn’t appear in the other two.

Millionshort.com: longtail research. Eliminates top 100 to 1 million sites. Catches less SEO’d sites. Essential for more complete research! See also MillionTall.com: only top sites.

Impact of spoken search: we’re now used to typing in a query. New software like Siri has changed that dynamic, along with the wording of the query. Google is having to deal more with natural language queries; people will word a query like a question. Compare the results of these queries: What is the meaning of life? The meaning of life? Meaning of life? Significantly different results for each query.

Are we as relationship focused as Google Now? It’s an app on your phone, which looks at calendar, gmail, search history, etc. Automatically populates your phone with what it considers to be pertinent information. Library users will get used to this – how can we make sure we’re as client-focused?

Find info from trusted sources: Bing only – ID outgoing links from a site or domain: LinkFromDomain. LinkFromDomain:msf.org ebola retrieves pages linked-to from Doctors Without Borders that mention ebola.

Data-mine Academic Search: search result frequency counts – ID the key authors, conferences, journals, orgs, keywords. academic.research.microsoft.com (Bing academic search)

Digging deeper in Google, Bing: getting into the deep web. Look for databases: {topic} database. Bing only: find pages that link to a filetype: contains:mp3, contains:xls

Find LinkedIn “hidden” updates: mouse over the “send a message” button, and it will give you options, including ‘view recent activity.’

Tineye.com: looks for similar images. Is anyone using your photos? Did they pick up on your press release?

Bing’s Image Match: ID an infographic from a report – look for other pages that have the same image.

Zanran: “google for data.” Web crawler looks for images on web pages, including charts and graphs.

Shothotspot.com: search Flickr and Panoramio for images on a location. Shows pictures people have taken that were geotagged.

Search the Internet of Things: thingful.net. Identifies places where people have tagged things in the Internet of Things.

Gwittr.com: learn about influencers and their influencers. Who they retweet, what links they share, what they hashtag, how much they share.

Leave a comment

Filed under internetlibrarian