Computers in Libraries 07, Wednesday

Greg Notess, publisher SearchEngineShowdown.com, “Book Search Engines Update” (A305)

See his book search page (http://searchengineshowdown.com/booksearch) and his presentation (http://searchengineshowdown.com/booksearch/cil07)

Full text searching of books is for:

• Different information source
• Searching, not reading (usually)
• Verify citations and find mentions, such as first mentions of a term
• Consider other potential uses

Scanned/converted books:

• OCR quality varies
• Full content scans
• Electronic files conversions – may lose initial letter or special characters (such as apostrophes)
• Huge collections of data
• Multiple editions of a work because different library scan the same item

Book Search Engines:

• Amazon’s “search inside the book”
• Google Book search
• Open Content Alliance
• Internet Archive Text archive
• Live Search books
• Individual publishers’ initiatives
• Open Web

Amazon and A9.com

• When in Amazon, go to the books store to search
• A9, use the books box to search only books
• Search inside is different from look inside
• Currently published books
• Including reprints

Google books is “to help you discover books, not read them…”

• Books.google.com
• Scans of books or electronic copies from publishers
• The have agreements with publishers, and if no agreement, they are not included unless it is in Google Library.
• Google Library scans books from libraries
• Google considered items in copyright since 1923, even if it is not.
• They have three levels of viewing:
• Limited access
• Snippet view
• Limited view
• Full view
• No preview available
• One problem is older fonts were not made to be scanned or read online.
• Google provides links to: the publisher, stores and shopping sites, and OpenWorldCat and 14 other union catalogs.

Open content alliance:

• Internet archive, Yahoo, O’Reilly, Microsoft, and others.
• Some is in live.com books. Do a search, click “more” under tool/tab bar, then select “books” and it will search just books
• In Internet Archive, go to the “texts” section.

Flip books format is used by Open Library openlibrary.org

If you want newer, copyrighted works, start with Amazon, then try Google.

In the open web, see Project Gutenberg (Gutenberg.org) and online books (digital.library.upenn.edu/books) and many other hidden spots. To find the hidden spots, search the title as a phrase search, and maybe phrase search from content. You can look for lists of books my searching for title with no space and by searching for the authors’ last name.

See his presentation to create google and yahoo searches to find ftp sites of books.

Publishers sites may have enough info to get you what you need, may show where it is full text, or actually provide some electronically.

Computers in Libraries 07, Wednesday

Wisnewski and Cervone, “Federated Search: State of the Art” (A304)

Federated searching protocols: Z39.50, SRU/SRW, OAI, and XML gateways.

Most federated searches allow for simple and advance searches, combine results sets with deduplication, and some statistics.

Advanced features for federated searches: integration with other software, such as courseware and RefWorks; context sensitive linking.

Top players: Ex Librirs, Webfeat, all WebFeat partners, Total WebFeat.

Proquest CSA has metasearch by topic, so you can search multiple databases based on the subject (such as humanities).

Ovid SearchSolver: based on Muse technology, is hosted, and results are post-processed.

Medline has faceted browsing. (check it out)

Ex Libris (loves) Endeavor, and creates MetaLib, SFX, and Primo, all at www.exlibrisgroup.com/(name).htm

Metalib 4.0 has faced browsing, and its interface will eventually be replaced with Primo.

Aquabrowser is a unique interface that is part clusty type linked terms and part catalog.

III has Encore, which includes tag clouds which includes phrases (LCSH) as well as individual words (keywords)

WebFeat Express is a hosted version of WebFeat Prism price “affordably” at $8000. It has limited conform/adaptability compared to Prism.

Who else is in the game: AGent – Auto-graphics Inc and many others.

Other companies are marketing federated searches to large companies outside of libraryland to help their knowledge management. Inktomi (old name) has Universal Repository Interfacing, Siderian, with its Seamark Navigator, and Endeca.

Trends:

• Number of vendors is shrinking
• Migration to XML from Z39.50, tho Z30.50 still prevails
• Integrating content with external systems such as WebCT and RefWorks
• Increased use of visualization and clustering
• More off-site hosting possibilities
• Data pre-processing options
• Free versions with Google scholar and Microsoft Academic Live.

Their ending question is much like mine; why do a federated search when the big search engines have scholarly versions.

Computers in Libraries 07, Wednesday

Marydee Ojala, editor, ONLINE Magazine, "Searching, Finding, and the information professional” (A303)

Check the infotoday.com blog for more information.

Agenda:

1. Searching
2. Finding/findability
3. Search techniques
4. New search technologies
5. Nontextual information
6. The future

Searching:

Information professionals/Librarians “enjoy the ‘thrill of the chase.’ They understand how information is structured and where it is likely to be. They know the intricacies of search engines, and often exchange search tips” and warnings of tips that stopped working (~synonym on Google).

On the flip side, sometimes information professionals forget to stop searching, or don’t know when to quit a search. They also have problems overlooking things due to information overload.

There are so many places to look and so many tools/resources to use that it takes much longer to do a search now.

Finding:

Clients want the answer and do not care about the source. They have a “non-Boolean mindset.”

Info professionals want to find info, as well as seeking, but they care about the source too.

Findability:

This is the flip of finding. Website creators organize their sites, and do other optimizations, to get themselves to the top of the list for what ever searches they want to be on.

Premium content providers' (EBSCO, etc.) findability is expanding into other markets, including pay per view thru a search engine. They are also working at becoming more intuitive and looking more like web searches.

Search is pervasive, but it is unstable and gets gamed. The content has quality issues and our clients need to be educated about these.

Search Techniques:

Traditional techniques like Boolean, Pearl growing (start with key and grow out of it), building blocks (build off of your key ideas/terms), and successive fractions (break large search into smaller parts.

Web techniques use squishy Boolean, where the search engines coding tweaks my Boolean terms. Also need to use invisible web, metadata, and other tools.

New Search Technologies

Personalization:

Web search engines know who you are and what you want, which is great for individuals, but not so great for info professionals who want non-biased, non-personalized results. (It also has problems with people who search a wide range of things for multiple people and/or ideas.

Optimization:

Web search optimizer (SEO) community has a lot for Info professionals to learn from them.

Set your search result preferences to 50 or 100 because the top 10 are more likely to be gamed.

Semantic Clustering:

Contextual searching that tries to match words with multiple meanings to the correct meaning you are searching for. One way to do this is with visible clustering, such as Clusty and Factiva.

Automatic Indexing:

It is good in theory, but works best with human oversight.

Metadata:

It is discredited as a means to drive traffic because of abuses, but it is great in a controlled environment.

Different databases:

Each search engine has a different database. Premium content searches are different from a database producer and an aggregator.

Invisible/Hidden web:

Many formats are no longer hidden, but there is still much that is hidden, especially historical pages and sources, and limited access things.

Nontextual, nontraditional information:

Searches of audio files, video clips, images, blogs, groups, and second life are rather weak. A large problem these face is how to describe or define these things. Media is more often found by viral marketing, told or read about it, not searched for it.

Displaying nontextual information:

EBSCO and Gale’s use of Grokker, and Factiva’s tag clouds are ways to show results in non textual/list ways.

The future

Worst case scenarios:

• A controlled information environment
• Consumerism and entertainment trump research
• Industry consolidation diminishes available information
• High price doesn’t guarantee quality data

Best case scenario:

• Intuitive interfaces
  
• No licensing wars
  
• Information is accessible and available
  
• Producers are profitable
  
• Searchers are satisfied
  
• Searching and finding coalesce

Computers in Libraries 07, Wednesday

Gary Price, “Keeping one click ahead: Best of ResourceShelf.” (A302)

His presentation is located at http://tinyurl.com/2qe8tt.

• Skylineglobe.com is a mashup of digital globe (so far US), maps, and live traffic cameras.
• Terrafly.com meshes aerial images with searches about that area (local businesses, schools, towns, weather, etc.).
• DocuTicker is a ticker tape of government documents.
• Citizendium is similar to wikipedia, but it requires people to sign-in when they edit/add information.
• Webharvest.com archives government websites and documents, and allows you to keyword search the database.
• Footnote.com is part free and part pay.
• NewspaperARCHIVE.com is mostly free for K-12 schools and public libraries. You just need to fill out a form and send it back. It is full text and full image articles, and has many special collections on topics.
• Loki uses wifi to say where you are and give local information.
• US dept of Commerce commercial guides are pdfs available for free at http://www.buyusainfo.net/adsearch.cfm?search_type=int&loadnav=no.
• At the bottom of the page is a list of web 2.0 apps and tools, but he cautions that people are not using web 2.0 tools as much as media reports.
• Tucows.com has a section of web 2.0 apps.
• Farecast.com datamines airfare lists to help you determine the best time to buy airplane tickets.

Computers in Libraries 07, Wednesday

Mary Ellen Bates, “30 Search Tips.” (A301)

1. Simply Google
1. Overview of all of Google’s features, sites, downloads, etc.
2. http://www.usabilityviews.com/simply_google.htm
2. askx.com
1. ask.com’s new interface
2. left column stays same with search suggestions.
3. Middle is your search results
4. Right column is quick views, pictures, etc.
3. Searchtheweb2.com
1. Can list by popularity or by long tail
4. Google search history
1. Can limit search to what searched before
2. Google.com/searchhistory
3. Can turn search history on and off
5. consider the Q&A services
1. answers.yahoo.com
2. qna.live.com
3. askville.ask.com
4. linkedin.com/answers
5. Use all with caution!
6. use search engine’s quick-answer features
1. ask.com’s smart answers
2. google’s onebox
3. yahoo’s shortcuts
4. MSN/live.com’s instant answers
7. Squidoo
1. blog on steroids
2. makes a more interactive with polls, etc. into you blog.
8. build your own search engines
1. rollyo.com
2. yahoo search builder
3. google co-coop
4. live.com’s macros
5. gigablast’s custom topic search
6. – can add to your website to get students to you Google your way
9. Yahoo search subscriptions
1. Searches periodicals, but a pay-as-you-go.
2. Search.yahoo.com/subscriptions
10. Kosmix.com
1. Is a vertical search engine that gives search results in clusters and related topics
2. Has for politics, finance, and health/medicine, and a few others
11. NationMaster.com
1. Datamining tool covering data from NGO’s.
2. It is also a data visualization tool; it makes charts out of data from reports so it is easier to read/understand.
12. Yahoo’s mindset feature
1. Mindset.research.yahoo.com
2. It has a slide to show where you are between researching and shopping.
13. MSN’s cool synonym-suggestion tool
1. Snurl.com/zqyd
2. Intended to help advertisers find search terms to advertise
3. Is great for finding other terms to search by
14. MSN’s misspelling-suggestion engine
1. Snurl.com/zqyl
2. Gives the most common misspellings of a word
15. Consider 37signals.com
1. Great collaborative resources
2. Does project management, to-do lists, file sharing, time slips, etc.
16. Ask.com’s maps
1. Gives driving AND walking directions
2. Takes topography into accont
3. Maps.ask.com
17. use Exalead’s NEAR/n operater
1. (solar OR sun) NEAR/3 power
18. Live.com’s academic search
1. Academic.live.com
2. Great results page
3. Slider bar for verbosity
4. Endless scroll of results
5. Many search options
19. Compare search engines
1. Comparesearchengines.dogpile.com
2. Jux2
3. Ranking.thumbshots.com
4. Twingine.com
1. Don’t use for searching, but use for evaluating and for graphic representation of search engine coverage (and lack of overlap).

20. Data visualization arrives to the value-added online services
1. Factiva’s discovery pane in Search 2.0
21. Touchgraph
1. Finds relationships among URLs
2. Uses Google’s “similar pages” function
2. Finds related books in Amazon by using subject terms.

21. Gigablast
1. Limit to multiple sites
2. Gigabits clustering
3. Great complex searching and ranking tools
1. Have to use the “Help” file for syntax

2. Snap search
1. Snap.com
2. It gives a preview of links in second frame
3. Has a firefox plugin
3. GooFresh
1. Limits google searches to only sites updated or added recently
2. From Tara Galishain
3. www.researchbuzz.org/2003/09/goofresh.shtml
4. check out podcast lectures:
1. Yale
2. Princeton
3. UC Berkeley
4. Stanford
5. Johns Hopkins
6. And others (and there is/are podcast search engine)
5. SearchMash
1. New interface for google
2. Does regular search, as well as images, blogs, and videos searches
3. When do “more results” it adds the next 10 to your screen
6. Google’s topic search
1. Google.com/coop/topics/?show=all
2. Information pros can add to it and make it better
3. Searching in it limits what you want to do with the results
7. Pagebull
1. Metasearch tool
2. Entirely visual (screen shots of all search results)
3. Good for right-brained searchers
8. Google’s librarian center
1. Blog, newsletter, and teaching tools
2. Tips of the trade
3. Librariancentral.blogspot.com
9. OneLook
1. Reverse dictionary
2. Onelook.com/reverse-dictionary.shtml
3. You enter your definition and it says what words it might be.

Computers in Libraries 07, Wednesday

John Van Oudenaren, Sr. Advisor, World Digital Library, Library of Congress, “The world digital library initiative.”

This is not a mass book scanning, but a representational “significant original materials representing all the major cultures from across the globe…”

It is in large part a content acquisition project. This project works with other national libraries such as Brazil, Russia, and Bibliotheca Alexandrina, with UNESCO, and with the technology community (Google, Yahoo, Apple, and others.)

LOC proposed the WDL in June 2005. WDL prototype is to be unveiled at UNESCO General Conference in October 2007. Goal is to complete the WDL plan and fully launch in September 2008. LOC has spent the time between 2005 and 2007 raising funds and getting agreements with partner institutions.

Three pillars of the project: content acquisition, construction of sustainable network for production and distribution of content, and the web site: www.worlddigitallibrary.org

One objective is to increase the non-English items, especially archival ones, on the internet.

“Key Objective: Present cultural content in a way that appeals to (and thus will be used by) the new generation of Internet users, both in the United States and internationally.”

The WDL will be multilingual: English, Arabic, Chinese, French, Russian, Spanish, Portuguese. The scans will be in the original language, but will have translation features too.

They plan to have some audio and video, educational content, some social networking features (didn't say which ones), and adjustments for developing-country conditions (low bandwidth and mobile devices.)

Computers in Libraries 07, Tuesday

Steven Cohen, “What's hot with RSS!” (A204)

His presentation is at http://stevenmcohen.pbwiki.com/CIL2007, and he uses pbwiki over powerpoint for presentations.

• He prefers Google reader. One feature is to subscribe to others shared items from their feeds (thru Google reader).
• Windows Vista now comes with feed list (titled “news”) on its desktop.
• Internet Explorer 7 has feed reader built in.
• Libworm.com – librarian rss search engine. You can save your search as an rss feed too.
• Page2rss.com – changes websites into rss feeds so you can track a URL.
• News feeds are not that good anymore. Some times the email alert, like from Yahoo, provides a very different amount of results from the RSS feed.
• Techmeme.com – ranks blogging by topic. It shows what is hot.
• OpenCongress.org – kind-of mash-up about bills, issues, and politicians covered in the news.
• Dockets.justia.com – federal district court filings & dockets.
• University of Oklahoma Libraries has great RSS feeds from their catalog. It tells you when they add a new item on your desired call number. They also have feed for added electronic sources, but that could also be used for what has changed (new years added, trouble accessing, more journal coverage, etc.)
• LibraryThing has a lot of feeds.
• Twitter – you can set up a feed for every friend, so as they add things, you get notice in your rss reader.
• Rss2twitter.com – add an rss feed to your twitter account.
• Tumblr.com – a blog of links.
• With some work, twitter could be the next stage of rss readers

His top 10 – or 12 – favorite tools

• Internet archive – archive.org (also great for reference)
• Snapper – firefox add-on (https://addons.mozilla.org/en-US/firefox/addon/2703) which allows you to do screenshots of parts of webpages, and saves it as a .png file.
• Browster
• Bugmenot
• tinyURL
• googlegroups – groups.google.com
• citebite – provides links to part of a page so you can someone to a specific part of a page.
• picnik
• missing-auctions.com
• twitter
• meebo
• pbwiki

Computers in Libraries 07, Tuesday

Meredith Farkas and Michelle Boule, "Learning with blogs and wikis." (C203)

Their presentation is at http://michellemeredith.pbwiki.com/

It was a great presentation of how blogs and wikis can be used in teaching and learning, based on their experience leading "Five weeks to a social library."

Computers in Libraries 07, Tuesday

Nicole Engard, “Project planning the 2.0 way: Using blogs & wikis” (D202)

Her blog is at http://web2learning.net/ and will have her canned presentation and she may add screen shots from her wiki.

Instead of having meetings (either some people are missing or there are fewer meetings), and trading emails and phone calls, use a blog, because all can access it, when they have the time, and peoples’ activity is tracked by their number of comments/postings. Also all statements are in a common area so don’t miss out on reply to sender vs. reply to all. Comments are also archives and much more easily searched.

Each project has a blog and each staff member has the ability to setup a blog and interact with them. Everything still gets time stamped, and it cuts down on emails.

When the project is complete, put the documents on a wiki on the intranet. It is web-based, full-text, archived, backed-up, has history of edits, and open to all for when each needs it. Documentation on projects (goals, follow-up plans, etc.) are all in a common area for all to see and use. They also put a shared library calendar on their wiki.

They have a help button on each page, so when there is a problem, the error report comes from/with the page that is having problems.

Engard then showed a live use of her blog and wiki interface.

They used WYSIWIG Pro for creating/posting, and it costs $40 for non-profits.

Wiki’s have history so even if someone messes thing up (even if they attempt to destroy information) it can be restored by the administrator.

They can collaborate on training and guides, because the wiki is editable with comments. And you do not have to remember where you put it because the wiki is searchable.

Q&A

- How do they do project archiving & change it from blog into wiki? They edit a project and archive it. Then they have to change it over to a wiki.

- It is their creation and it is based on php and mysql. Joomla is an opensource/free software that you could use instead of doing your own code.

- You can also use wordpress for the blog, then download it for your server if it works for you.

See http://web2learning.net/learn-more

Computers in Libraries 07, Wednesday review

When I get back to work on Friday, I'll try to get the rest of my notes up for the conference, but I wanted to get out today's review while it is still fresh. Today's opening session was ok for its topic, but the presentation of it was not great. The speaker seemed to fall victim to the tyranny of PowerPoint. It was about an LOC project with UNESCO and many other national libraries to digitize culturally relevant materials from around the world. So that was the worst presentation of the day. I went to the A sessions on searching and they were all great; good topics, good tips and strategies, and good presentations. Some of what I learned was:

• The increase in search tools means the time it takes for a thorough search is taking longer. A corrollary is you have to know when to end your search because there is just so much out there.
• Federated searching is improving with tabs, visual searches, and tag clouds, but there are big challengers in the academic search engines, such as Google Scholar, which need to be considered.
  
• E-book searches (Google Scholar, Amazon, etc.) are getting good and are worth using for research.
• Gary Price will do online or teleconference brown bag presentations!

In summary, the search aspect of librarianship is only going to grow as searching becomes more complicated and deeper.

Computers in Libraries 07, Tuesday review

Many of the sessions I attended today were practical; they were either "how we did it" or "use these tools" sessions. These are great things, but after the statistics in OCLC's and Pew's presentations on Monday, I'm taking these sessions with some skepticism. If librarians are far ahead on the digital application spectrum, are our audiences really wanting these web 2.0 services? If some do, is it enough in either size or strength of desire to spend the resources necessary to correctly implement social software?

My tentative answer (only 2 of 3 days done and not much time to reflect) is yes. Allowing students to help create research tools thru the use of wikis or similar tools creates buy-in. Having a blog, with RSS feed per Cohen's request, keeps users informed of news and situations, and it allows them to interact with the library thru comments, thereby humanizing the library.

So what did I get from today?

• Use WorldCat.org and its downloadable search box to help spread the news and use of library holdings.
  
• Use RSS feeds not just for new items but also for trouble connecting, more content, and similar news that may be small, but helps keep the library in front of your users.
  
• You should use CMS because there are many good options for content manangement, including blogs and wikis, which take little or moderate training, respectively.

So far I have learned some things, confirmed much more, and generally enjoyed my self professionally.

OCLC breakfast @ CIL

OCLC had a great breakfast where they gave updates, news of upcoming products and services, and table time to talk with other librarians and with OCLC reps. Some of the news and updates were OCLC's three free services: Worldcat.org, openworldcat, and Worldcat registery (see www.oclc.org/worldcat/web/). Worldcat.org is working on providing social web functionality to the WorldCat catalog, such as tagging and patron written reviews. It also includes a search box that you can download and put on your library or personal website, or blog, to let others search all the records in WorldCat, tho not all of the holdings.

Openworldcat is OCLC working with booksellers and search engines to get more book records in search results. WorldCat registry is a list of participating libraries, museums, and the like. These two didn't interest me as much as Worldcat.org, but they still sound useful.

This summer or fall OCLC plans to have 12 RLG databases transferred over to FirstSearch. Along with the databases, OCLC is starting OCLC Terminologies, which is controlled vocabulary of art from the Getty. Finally, good are vocabulary! And to aid librarians in learning and using OCLC's resources, OCLC is starting Web Junction (webjunction.org/learningcenter). It will be a site were librarians can interact and learn together, and an e-learning resources for librarians for continuing education.

It is good to see some part of libraries' catalog are moving towards web 2.0. (Are you listening III and company?)

Computers in Libraries 07, Tuesday

Session: Using a CMS to build community: Rhumba with Joomla (B201)

(see http://www.joomla.org/ and http://www.joomlainlibrary.com/ for more information)

The South Carolina State Library, where the speakers are from, uses joomla for its website content management and this is how the set it up.

Joomla is free and open source, easy to set-up and use, separates content from form, portable, and extendable (over 1000 add-ons). There are other large organizations that are using joomla, such as Porche Brazil, SC IT Directors Assoc., and United Nations RIC. If you want to see some evaluation, www.cmsmatrix.org/matrix compares 10 content management systems, so you can see how joomla compare to other cms software.

You can use Joomla as a wrapper around your catalog, or you can imbed your catalog into Joomla/the website.

Redesign process

The stages of the process and some of the lessons learned.

Phase 1:
Did online (survey monkey) and in-house surveys to get broad inputs.
Reviewed current content
Need people not familiar with the subject area to review the so get fresh eyes.
Have a project manager.

Phase 2:
Interface design and review
Process stalled with new director.
Wedsite designer did not understand library’s issues and came up with unworkable website.

Phase 3:
Explored CMS options
Joomla learning curve
Translating graphical interface to Joomla templates
Migrated or rewrote content

Phase 4:
Give staff some time to review new site, but go live soon to force them to use the new site and review it. Evaluate it, the plan for growth and refinement.
CMS makes it much easier to evaluate the depth of your content. It also helps by making a site map.
Send the web administrator away so people use it and talk amongst themselves so they learn it.
All is done/based on CSS so it is 508/accessability compliant.

Before this process, they had a homepage committee, a PR committee, and a web administrator. Now they have 2 web managers, 25 authors, and 325 registered members with 130 yet to be approved.

Joomla template has three parts: index.hph, template­_css.css, images, and templatedetails.xml

Staff can add content thru a browser add-on that is WYSIWYG, and which provides pre-defined content formatting. Joomla also reinforces accessibility guidelines so no matter who is writing, it will be compliant.

Joomla seems to be a great tool, but both presenters warned that it has a steep learning curve and the integration was often frustrating, but worth it. So if you have time but no/little money, Joomla could be for you.

Computers in Libraries, day 2

Today I am using my computer for notes, so I will probably have more information on each session that I attend. I may even upload my notes for each, if they are not too long.

Computers in Libraries, 2007, day 1

Today was the first day of Computers in Libraries 2007 conference. A person from OCLC gave the best and worst presentation: she had a lot of hard numbers on internet usage and attitudes for the U.S. and some other countries (the report is due soon) but her fonts were hard to read from the front of a very large room. Some other things I picked up are:

• Users want to interact with the internet: allow them to comment, rate, tag, edit, or do something so they feel ownership and a part of a human community of your library. PLAN before you do web/library 2.0. Most social software is easy to setup, but it needs a lot of planning before you start so you know where you are going with it and how it supports your libraries mission.
• Don’t try to do or use every web 2.0 services, just a few which meet your personal/professional needs or the needs of your users.
• When you advertise it, don’t says “we have a blog” but rather say “we have a new way for you to keep up with the new databases, books, etc.”
• Librarians read and use the internet much more than the general public.

That’s all for tonight.