They’ve got it too easy so don’t develop assessment skills
They’re supposed to be savvy, hip, and oh so good at the web to find what they want. But a report from London University says that while young people have an easy familiarity with computers, they rely on the most basic search tools and do not possess the critical and analytical skills to assess what they find on the web. Librarians whom ACCESS has talked to said “we know that already”. So get into the “I told you so mode” when you read the report Here.
Shock waves whiz around the world…
The writing was on the wall. But it still came as a shock to publishers and many others when Harvard’s Faculty of Arts voted to give the University a world wide license making each faculty member’s scholarly articles freely available and to exercise the copyright in the articles provided that they are not sold for a profit. The result will be that anyone, anywhere, will have online access to the scholarly work of the Faculty. Will your university follow suit? More here
…Publishers criticize new NIH mandate on journal articles
Under the previous voluntary policy, NIH-funded researchers were requested to submit an electronic copy of their manuscripts when accepted for publication in a journal. That has changed with President Bush signing a bill directing the NIH to provide public access to findings from its research. Now, say the publishers, this mandatory policy gives the NIH publisher property interests and undermines publishers’ ability to exercise their copyrights on articles they publish. Is America showing the world of scholarship the way? Two articles Here will help you make up your mind. .
The library automation industry in a many headed beast
New companies appear, old companies enter ‘strategic partnerships’ and die, they chase each other to add new technologies… so you could be forgiven for thinking that they all do the same thing and deliver the same level of service. But you would be wrong, as Marshall Breeding’s 2007 survey shows. Librarians’ satisfaction with their library system varies widely. Who is doing well and who needs to pull up their socks? Turn to this page.
Can there really be a rival to Google?
And could it really come from a university? If you’re the University of Manchester then the answer is yes. The free service adds thousands of documents to the Intute service. Researchers can access papers from research databases within universities which have been harvested from trustworthy sources guaranteeing high quality, something which public search engines are unable to do. This page has more.
Library revives lessons of Confucius
70 students in red robes recite a proverb saying “We should get up at dawn and clean up the court to keep order both inside and outside.” Before you fire your cleaners and hand out brooms to library visitors, the library in question is in China where, well, they do things differently a lot of the time. But if you’re tempted to initiate some community service in the stacks, read here.
Is anybody not doing it?
Digitizing vast collections of books that is. The French are doing it, Europeans too, Google, Microsoft… So what’s so special about The Million Books Project? First, libraries in four countries are collaborating. Second, many of the books are in languages that aren’t English. For example, 50,000 in Telugu, 40,000 in Arabic. That’s quite a feat. Of course, there are some glitches: many of the titles are still in copyright so can’t be offered for free. Find out what can Here.
English language citation hegemony over?
They’re taking over the world. And sometimes that’s a good thing. The Web of Knowledge has just gone bilingual with the hosting of the Chinese Science Citation Database. Information will appear in both Chinese and English and searches can be made in both languages too. China published 70,000 papers in 2006 ranking it 5th in the world. Need more mind-boggling figures? Get intoxicated on this page.
Is social networking just for teens and tweens?
When you think about it, sharing content, collaborating, creating communities has been around for ages. Research shows that visitors to the main social networking sites stay connected for about 90 minutes. That may be more time than they spend in their university library in a week (we won’t say month!). Where do libraries fit in to this new social order? Nine experts tell us how starting this page.
Groxis, the company, was formed six years ago by founders who felt with so much information in the public and private domain, we need the ability to find needles across multiple haystacks and discover relationships that are not immediately obvious. This need developed into Grokker, a tool for visualization and clustering of search results. ACCESS spoke to Groxis CEO, Randall Marcinko in Bangkok recently, to find out if we’ll all be Grokking after Googling.
What is and isn’t Grokker?
Grokker is not a search engine. However, we are agnostic and work with all search engines. Grokker sits on top of multiple sources and provides federated results. Instead of delivering a vertical list of sometimes thousands of results, Grokker assembles and normalizes the search results, grouping or clustering them both textually and visually. The results are a new search paradigm, allowing users to visualize connections within groupings and better analyze search results. End users most frequently look at less than three screens from the thousands of returned search results. Using Grokker, users immediately see the cluster(s) of greatest relevance, and drill down, only within the cluster(s) that matter to them.
Is Grokker unique or just the best of the bunch?
I think it’s both. There are many products which do one or two of the four or five things that are mainstays in the technology and the information market.
Grokker organizes and provides a visual map of federated search results making it easier to discover, explore and organize information. The maps use size, shape, colour and other attributes to present information in a dynamic contextual setting. Clicking on an item in the map presents additional information.
In the next year we will see many interfaces of varying interest to end users, scientific researchers and others. Grokker’s software developer’s toolkit enables developers to build various types of interfaces, whether heat maps pie charts or graphs Visualization is critical but not every industry or researcher needs the same flavour.
Is Grokker visualization customized depending on the customer?
Until recently there has been little customization - minor colour and functionality changes only. We are opening the door to different looks and feels to accommodate different industries. An epidemiologist might prefer a heat map to recognize symptoms indicative of a pending epidemic; while a chemist will need a different model to best visualize their problems. In the next six months you’ll see many new looks and feel to the database
Can customers choose which visualization and you provide the customization?
Yes. We have an open API so clients can customize themselves if they wish. More often, subscribers ask Groxis to build to their specifications. We’ll have at least 25 out of the box UIs this year.
What’s the connection with Stanford University?
Mike Keller, the Stanford Library Director, is on our Technology and Strategy Advisory Board. Stanford has been a long standing customer. Mike and our other advisors have contributed significantly based on their needs and unique experience and background.
Does Stanford serve as a test bed for each new release?
Not all released features pertain to Stanford. However they have been very generous with feedback that has been extremely useful.
What’s the link with Muse Global?
Grokker has traditionally connected to a small number of sources, applying clustering and visualization. More of our clients are choosing to search across a wide range of content sources. We are federating content from more than 10,000 different sources. We have established an OEM relationship with MuseGlobal, who we believe is “best-of-breed” in this space, just as we lead in our technology. This partnership complements Groxis’ own federation and related technology.
How does it all fit together?
The end user need not bother about the components. We supply a product that uses the necessary components, assembled for our clients’ needs. Federation is a layer within Grokker. A Grokker user selects their sources, within their customized environment, executes their search and then works with the clustered and visualized results. As necessary, Groxis builds in the connections to harvest the source data.
Are you working with library management systems?
We work with almost all the OPACS in order to harvest data that supplements traditional database search results. We also work with all the major aggregators and publishers too. Grokker seamlessly embeds connectors as needed and handles the appropriate authentication from client to for-fee data sources.
Do any library management systems include Grokker in their products?
We are talking to many library management systems regarding specialized ways in which our software can enhance both end user results and their display. It is much more compelling for the end user to know the catalogue availability of an item and not just its bibliographic citation.
For Grokker you specify which sources you want to search – databases, websites and so on. How about the entire web?
Groxis connects to the entire web through any search engine of our client’s choosing. One might choose to use the Google Search Appliance via Grokker while also federating results from Medline, BIOSIS and Current Contents. The results are clusters of content containing both pertinent results from proprietary databases and the open web
Describe a typical academic Grokker user.
A university who purchases Grokker will frequently choose to search 10 to 20 in-house databases. They often access 4 to 6 aggregators, have an OPAC, research partnerships and data sources at remote locations and need to combine these with results from the open web. They need to search across all or a subset of these sources on similar subjects. They want to see clusters containing content from each selected source. The clustered results are then visualized, textually and visually, in the UI of the client’s choice. Sources are vast with no additional source charges. The university may also choose to implement secondary processing where Grokker can indicate holdings information on each record. Document delivery lookup or other secondary reference lookup could also be included.
Who puts it all together?
Most typically the client requests Groxis to assemble the entire product. However, with more and more open APIs that we are adding to the system, the client will have as much flexibility to “do it themselves” as they might wish.
Randy Marcinko, CEO, Groxis
You’re talking to iGroup about supporting Grokker in the region. Will it be them that does this integration?
With iGroup, given their technical strength and strong customer support, it will be a collaborative approach. Groxis will support iGroup as needed. iGroup knows their customers and nuances of the region. In some instances Groxis will add to the product line and customize our joint offer to add functionality and localization best-suited to each country and for each language and culture.
You work with Gale - Cengage Learning - to embed Grokker into their PowerSearch platform. What’s that about?
When a Gale customer performs a search via PowerSearch Plus, they are able to view the returned results clustered graphically or textually. Perhaps even more exciting, all of Gale’s users are also allowed to use Grokker to federate search results from other vendors with whom they have subscriptions. This is extremely powerful for the library and end user alike. The end user is presented with a suite of services that enable them to maximize the value from all of their subscription content.
Is there only one version of Grokker?
Currently there are fifteen components. Most customers choose a subset of these components that become their version. In today’s world, we must be flexible. We do not want all clients to be forced to buy the same solution. We sell what the client needs and can afford.
Will we be seeing Asian language versions for customers in our part of the world?
We are already launching localized interfaces in various Asian and other languages. Groxis has a Spanish interface and a French interface which are very user friendly. Later in 2008, Groxis will also be adding the ability to search on Chinese, Japanese, Korean and Thai databases (in Chinese, Japanese, Korean and Thai) to cluster the results and to label the clusters in the source languages. We will soon add Arabic and Hindi.
Can you mix languages when Grokking?
It’s a challenge to be able to translate the meaning of search queries and search results across multiple languages. It is something we’ll look at in the future. We might include query translation at some point, generating multiple sets of results, one for each language.
What’s the greatest advantage of using Grokker?
Traditional keyword search generates a long list of results and can be extremely powerful. But it’s often insufficient to find the answer quickly and efficiently. Clustering takes one more quickly to the desired results. Consider the age-old paradigm of using a library. If I want to find out about a war that happened in China, I open the door to the library, walk to the aisle on China, then I walk down the aisle to find the shelf with books from the right era. Clusters are no more than an algorithmic method of accomplishing the same task, more quickly and efficiently and with greater precision and recall. Because Grokker provides both the graphical and textual interface, we provide maximum choice. We are starting to offer full database clustering where the searcher will be able to view the entire database in cluster representation. Or the end user might choose to use keyword searching coupled with full-database clustering. There is power in choice.
For more on Groxis and Grokker visit here.