Print

Print


Can I chime in that we already own a product which is designed to permit both private and public publishing of a knowledgebase articles.  It's even linked up to LDAP.

It's Footprints.  However, recently the knowledgebase portion of the product was turned off because of lack of use and the potential for breach of privacy.  Note that the privacy issues were not the fault of Footprints but more due to lack of implementation planning.

On Nov 8, 2009, at 7:57 PM, J. Greg Mackinnon wrote:

Ah... so perhaps to use a Web 1.0 term (where my brain seems to be stuck, sadly... maybe I'll just skip Web 2.0 and upgrade to 3.0 when it comes out...), you feel that we need a "Knowledge Base" web application that is accurate, succinct, and accessible.  I would agree with that. 

-Greg

David Todd wrote:
[log in to unmask]" type="cite"> Greg,

    This isn't a complaint.  It's a couple of observations and what I think is an opportunity.

    First, most of our clients would not benefit from following IT-Discuss.  In many cases, the discussions are too technical and are in problem-solving mode.  Most of our clients don't want to participate in figuring out how to fix problems, they just want to know that what they're seeing is a known problem and, if there is one, what the solution would be.  Searching the listserv for solutions would likely just confuse most of our clients because of the appropriate give-and-take that the list fosters.

    The Chronicle article was about the idea of using the experience of the community at large to let individuals help themselves.  While I think that's a good idea, that might end up being a case of the blind leading the blind (no offense intended -- just that sometimes people don't know what they're talking about, as in my response to Zoey early this morning).  I think a more beneficial system would be one in which the refined nuggets generated by this list would be deposited in a place where they would be more accessible and more succinct than they are in the form of discussion threads.

    The connection with IT-Discuss is that it seems to me there's an opportunity to use the discoveries and knowledge posted on IT-Discuss to seed a community information bank with validated information.

    A couple of our distributed support providers have pointed out the need and opportunity, and I think the Chronicle article points out that this is not just a problem here and that other schools are looking at ways to capitalize on it. 

J. Greg Mackinnon wrote:
[log in to unmask]" type="cite">What is the problem with accessing information on IT-Discuss? Listserv posts are archived and search-able... they even show up in Google Appliance searches (although not with high relevancy, perhaps unsurprisingly).

I am not saying that there is not a problem, just that we need to better understand the complaint. Rather than investing lots of time in making just one mailing list more easily accessible, I would like to see a solution that fixes the perceived problem for all active/relevant mailing lists. Do we need a better search front-end for Listserv? Do we need better global search integration for Listserv (such as a search widget that can be embedded into a portal, or into select UVM web template-based pages)? Do we need to replace Listserv with something more modern and flexible?

-Greg

David Todd wrote:
Good Morning, All,

The article below was from this week's eChronicle, and I thought it might be of interest to others on this list following the discussion here a couple of weeks ago that Bryan Fleming raised on 10/15. Bryan's question was whether we have a "primer of technical resources available on campus", specifically for IT support staff. We in ETS have been looking at ways to do that, especially now that we have redirected some effort to the ETS web page (thanks, Jonathan!). But Nick's point on 10/16 was that we have a wealth of information on IT-Discuss but no easy way to access it.

As this article points out, other schools are wrestling with this issue.

So, would mining IT-Discuss archives to seed a Wiki be a feasible way to start something like this for UVM?





*Colleges Try 'Crowdsourcing' Help Desks to Save Money*
The tech expert at the other end of the line could be you
Colleges to Try 'Crowdsourcing' Their IT Help Desks 1

Ben Weller for The Chronicle
By Jeffrey R. Young

At Indiana University at Bloomington, good help is not hard to find, but it's pricey. Questions to the 24-hour tech-support help desk cost the institution about $11.41 per phone call and $9.39 per e-mail message—and last year the help desk handled more than 150,000 inquiries.

All that advice adds up, and at peak times some in need of it are left waiting. So, in a few weeks, the university will try something different: letting computer users answer one another's questions.

Information-technology people call this "crowdsourcing," a buzzword that puts a positive spin on leaving the job of writing and editing to volunteers rather than hired experts. The idea is to open a Web site where students and professors can post their IT woes and share their solutions. College officials tell me they hope it will grow into a self-service support center for colleges nationwide—a kind of Wikipedia for campus computer problems.

After all, professors and students everywhere suffer from the same digital headaches: glitches in Blackboard's online grade book, corrupted Microsoft Word files on the day a term paper is due, problems checking college e-mail messages on their iPhones, and the like.

The new database—being discussed by leaders at a handful of universities—will let users rate the quality of answers and highlight which contributors are the most reliable. Anybody, not just Al the IT Guy, can play the part of techno-wizard.

Even talking about such a change represents a cultural shift for many IT departments, which were once the unchallenged technical experts on campuses. These days many more professors and students are savvy enough to work their way through problems on their own, and even to offer solutions the sanctioned experts haven't thought of.

In some ways campus tech-support leaders themselves are crying "uncle" as the variety of gadgets and software packages that students and professors bring to the campus grows beyond what any college can support. When I asked Sue Workman, associate vice president for information technology at Indiana, one thing that students might contribute to the new Web site, one of the first things she mentioned was a video-game console. "A student might say, 'My Xbox gave me this error; what can I do?' Or, 'My Xbox quit working; anybody know where I can take it to get it fixed?' It might not be highly technical, but just something that we probably wouldn't spend money on maintaining."

But can the it-takes-a-village approach that built Wikipedia work for the narrower world of technical documentation? Will busy professors and students bother to contribute? If they do, will their answers be accurate?

And even if such a system is workable and promises cost savings in the long run, who will pay for the initial development?

I'll put those questions in the queue for now, with the promise to get back to them in the order they were received. First, a few caveats:

No one is talking about hanging up on old-fashioned telephone support. Because technology has become so crucial to teaching, research, and managing colleges, an IT-support hot line is bound to remain as long as there are phones on campuses. The new system would supplement, rather than replace, the existing model.

And many colleges already have well-stocked databases of technical-support documentation online. Indiana has something called the Knowledge Base, with more than 15,000 articles on just about any technology installed on the campus (even one on connecting an Xbox to the campus network). Until now, though, only help-desk employees could add or revise articles, which means the resource is expensive to maintain and not always up-to-the-minute.

Though anyone can search the Knowledge Base (it gets about 18 million hits a year), the primary audience is help-desk staff members, who use it as a reference library when they answer calls. The new idea is that the expert at the other end of the line—or Web site—could be you.

*A Crusader for Crowds*

Dewitt A. Latimer is among the most vocal proponents of the crowdsourced model of college technical support. He's chief technology officer at the University of Notre Dame, where the help desk is open only from 8 a.m. to 5 p.m. "But our students don't stop learning at 5 o'clock, and the faculty don't stop teaching at 5 o'clock," he told me recently. And, unlike Indiana, Notre Dame does not have an online database of advice, even for internal use.

A couple of years ago Mr. Latimer attended a college-technology conference and had one of those aha! moments. The keynote speaker was Barry Libert, a co-author of We Are Smarter Than Me: How to Unleash the Power of Crowds in Your Business, who talked about how companies like Amazon.com were tapping into user recommendations to increase sales. "I was sitting there in the audience," Mr. Latimer said, "and I thought, This concept was very applicable to the higher-education space—it just needed somebody to recognize it and run with it."

Since then he's been running with it. He helped organize a session that drew a standing-room-only crowd at last year's national conference of Educause, the education-technology group, and began talk of forming a consortium to build a national IT-support database. That conversation will continue at a discussion session scheduled for this week at an Educause meeting in Denver.

Mr. Latimer also made the case, along with three other colleagues, in a bulletin published in April by the Educause Center for Applied Research. "While the private sector has quickly caught on to the power of crowdsourcing, higher education has not yet recognized those benefits as fully," they wrote.

The key to making the database work, he argues, is getting a big enough audience by attracting a large number of college partners. To nudge students and professors into action, he proposes that each participating college hold monthly contests in which the most-frequent contributors win a free iPod or some other geeky goodie.

As for accuracy, Mr. Latimer argues that Wikipedia has proved remarkably reliable for most types of articles, and that its contributors quickly weed out most faulty information.

That's an area that still concerns Ms. Workman, though. "My motto is, Bad information is worse than no information," she told me. She said that when possible, published items in Indiana's new open database will be reviewed by staff members. Contributions that check out will be given a seal of approval. Other entries will be 'Use at your own risk.'

What does the help desk think of the new service? "It's a love-hate thing," said Al Joco, a user-support specialist at Indiana, who said he has seen plenty of errors on Wikipedia and doesn't want technical answers from "some random yahoo." The expert-approved Knowledge Base should remain separate from the crowdsourced one, he recommended.

He perked up when I told him that several college-support staffs might contribute to a shared database, though. "The more people that contribute to a knowledge base," he said, "the better."

*
Who Pays?*

Early this year, the talks that started at Educause nearly led to the creation of a consortium to make a national help desk a reality. Negotiators even had a name for the project: "Hosted Integrated Knowledge Environment Project," or Hike. But the economy was in the tank.

Now nothing is certain. Notre Dame and some of the other universities participating in the talks said they could no longer afford the membership fees needed to run the consortium, or the staff time to develop the software to make the database work. Even the name has been scrapped, in hopes of finding something catchier.

Mr. Latimer still believes that a help desk that involves users is the only way colleges can hope to keep up with demand for support as budgets shrink.

Lately he's tried to reach officials of Yahoo, which offers a popular service called Yahoo Answers, with the same tools to manage user-generated questions and answers that the colleges were thinking of building themselves. It would be a "win win" situation, he says, if Yahoo would let colleges set up sections of that service, on which the company could sell advertisements.

So far, no answer from Yahoo.


Andrew Hendrickson
CAS, IT Administrator
UVM, College of Arts & Sciences
438 College Street #402
Burlington, VT
05405

802-656-7971
802-656-4529 (fax)


To submit a request for service please use: