Authority Unbound: The real upheaval lies just ahead


Peter Morville Writes, about Wikipedia,"The accuracy, objectivity, and currency is surprisingly good."

"Now, some old-fashioned librarians may claim that due to the pseudo-anonymous, multi-author nature of the Wikipedia, its articles have no authority. But they'd be wrong. Authority derives from the information architecture, visual design, governance, and brand of the Wikipedia, and from widespread faith in intellectual honesty and the power of collective intelligence."


Authority derives from the information architecture, visual design, governance, and brand of the Wikipedia....

Authority derives from the "visual design"? So, I just need to print everything in a nice 10-point Times Roman font and I'll have the imprimatur of authority? I don't think so. Similarly, the "brand" of Wikipedia has yet to be proven to be authoritative, and since there isn't really a well defined body working to protect the brand, I don't see that "branding" will be an important factor in determining the authority of the Wikipedia any time soon (unlike, say, the Oxford University Press, whose editorial boards strive to ensure the quality of the product of the Press). Finally, while the open information architecture ensures that errors can be corrected quickly, it also allows for errors to be introduced quickly, and for lots of cooks to spoil the broth.

Can we use this and apply it to our OPACs or maybe WORLDCAR?

WORLDCAR? Finally, the melding of catalog and automobile! What kind of mileage can we expect?

Oops. So much for verification.

But I'm quite certain that Berkeley Oldstyle really was created for the University of California, Berkeley. By Goudy.

Berkeley Oldstyle Book is the "book face" version, a little lighter than the regular Berkeley Oldstyle. (Cites & Insights uses both: Berkeley Book as a body type, Berkeley Oldstyle for bibliographic citations.) All of which is a little arcane for trusting and verifying!

Not 10-point Times New Roman. That's the New York Times (and umpty million lazy site builders, since it's the default in most browsers).

Actually, it's the Times of London (or, if you want something more authoritative, Times of London [which will only work if you subscribe to Britannica Online, otherwise, look up Stanley Morison]).

"Trust, but verify" was exactly the point of a recent Slashdot story on Wikipedia. Old-fashioned librarians disavow Wikipedia (often without evidence) just because it can be corrupted. But as the links in that story show, the days of librarians serving only 100% trustworthy information are long gone, if they existed at all.This is a pre-Internet, pre-open stacks, pre-Gutenberg mentality. Not even LII can offer links to guaranteed information, but for some reason librarians exaggerate their role in providing quality information to users. People use Wikipedia because it's easy, another good point. Librarians cannot insist their users rely upon what they have selected for the stacks. Heck, even Colin Powell says "All you need is a search engine."Given all of this, it is to us to educate and inform users about evaluation skills, not harp on known vulnerabilities with information formats.

I see where you are coming from. Pulling out my trusty straw man, it seems to me this is yet another example of where a triangular hierarchical organization is uncomfortable with a non-hierarchical democratic, 'spherical' organization. The people top of the triangle are trustworthy and smart, while the bottom of the triangle are not very smart or trustworthy, at least compared to the experts. So when the great unwashed write an encyclopedia, its not considered any good because no one trusts them. But as you pointed out, Wikipedia is pretty good - and in many articles just as good as a commercial reference work.

I wish you had focused how the editors and the community have pulled back from the original anarchy, and have created a virtual editorial staff who cleaning up old articles, implementing standards, doing fact checking, etc. Maybe some of these ideas that librarians and others have used for a 100 years are actually pretty good.

*For the record, I'm a big Wikipedia fan.

Speaking not as a librarian (which is just as well, since I'm not one), but as a reasonably intelligent user, I find myself relying on the adage in the title of this comment when dealing with Wikipedia.

Some of the articles I've seen support that notion: They include sources (either electronic or print) that I can verify, and the choice of sources may increase or decrease my trust that the article's not biased nonsense.

Some don't: There's no way to check what's being said.

The other night, my wife was looking for the author and full text of a famous short poem where she just knew the first line. After trying Yahoo! and Google searches with little success (as those indexes get huge, the results get strange), I thought, "Wikipedia couldn't hurt."

In this case, the Wikipedia page triggered my own memory--the poet's name was correct--and the full text was exactly what I expected it to be. Score one for Wikipedia.

Does that mean I'd trust Wikipedia implicitly on something more important? No--but it does mean I might use it as a starting point more often.

As to design: Not 10-point Times New Roman. That's the New York Times (and umpty million lazy site builders, since it's the default in most browsers). Heck, the New York Times runs corrections, so they can't be absolutely trustworthy! Now, Berkeley Oldstyle Book: That's a sure sign of absolute truth, since it was designed for UC Berkeley, where errors are not allowed. (OK, if you're going to use a site's design as the basis for trustworthiness, this paragraph makes as much sense as any other--and establishes Cites & Insights as an absolutely trustworthy source.)

Good points.We may also be more likely to fix something we are an authority on.As far as which drink is in which glass, I guess I'm still not sure, I guess they're all half full at this point, I'm waiting to see which one gets emptied.

Couple of points here. I've also been published by two of your publishers and a bunch of others and I have had my work checked - not only by proof readers, but by my peers working in the same area. So I think we'll need to differ on that issue, but I do fully take the point that you're making.The other point I'd like to respond to is your final question, and I think you've actually answered it. It's books and journals plural. A good information professional isn't going to trust one resource - if it's important information they are going to check a variety of sources to be as close to 100% as they can. It's only after doing that over and over again that you can start to put trust in a publication (or indeed an author!), based on experience. I also accept your point that errors can still creep in and I'd never say that any reference work is going to be 100% correct.However, since I would automatically check any data that I found in the Wikipedia against another source (preferably a trusted one, but if not, against several others) I really fail to see the point in checking it in the first place. I might be old fashioned here, but I like to know where I'm getting my data from, and if I'm after facts, rather than opinions, then I'm going to lean more towards trusting a source that I know (or can check out) than an anonymous comment.Of course, Professor So and So might well be wrong, or have their own bias, but it is my *belief* that they are going to be more accurate than some un-named and unknown source.It's a judgement call, and at the moment my judgement, and that of a lot of the librarians that I work with, is that it's a call too far.

I think a crucial point you make here is 'we have the option to fix it'. If we take 'we' in a wider context as all Wikipedia users, rather than simply librarians, it seems logical to me that we're more likely to want to fix something that we feel strongly or passionately about. The more strongly we feel, the more likely it is that we'll have our own bias in what we say, and simply tilt the balance back again in a yo-yo that never ends.

Of course, if I see a factual error - 1816 instead of 1861 for example, I'd be tempted to fix it. But I'd also want to check that I'm 100% sure that I'm correct, which means going to a source that I trust. (One could therefore argue, though I won't strongly, that all the Wikipedia is, is a collection of biases and facts cut and pasted from more authoritative sources, which again doesn't really appeal to me.) If I am going to check that source it's going to take time, and I'm not sure of the extent to which a busy information professional is going to justify correcting a resource they're not going to trust in the first place.

To further your final point, for me it's not a case of 'half full or half empty' it's a question of 'it's the wrong drink in the wrong glass'.

I agree that it's premature for academic researchers to rely on the Wikipedia. But you better believe that students and the general public will be using it more and more in the coming years.

And I agree that many Wikipedia articles are written by people with an axe to grind or a particular bias, but the same is true of many journal articles and books.

In fact, I've written some of both, and they've been published by respected publishers such as Neal-Schuman and O'Reilly Media and Emerald, and I've never run into any "fact checkers" who questioned or corrected the factual content of my original text.

So, why do librarians trust books and journals so much?

Good points. Though I think some part of authority, at least online, comes fomr design. I don't know why he thinks so. I hadn't thought about it before, but I think of sites with better design as better sites. A well designed site is just one part of its authority, but I must admit it's a part for me.It's good that most librarians notice the bias and problems with these forms of authority, that's why we're here. None of us should happily accept the information where we find it. The good thing is with a wiki we have the option to fix it. We should also have the tools and experience to know where to find the best information. At this point that's probably not Wikipedia. It's still amazes me that Wikis work as well as they do, sure they have their problems, but I'm a glass half full guy at this point.

But what if your dean's PPT presentation is made available online, and someone who doesn't personally know him encounters it at the end of a Google search? I'll bet the MS Comic Sans would seriously detract from his authority (or his perceived authority if you prefer).

I read an interesting account today from Nicholas Carr, talking about the amorality of Web 2.0 and Wikipedia. He lists a couple of examples where the Wikipedia entries are incorrect, badly written, poorly composed and essentially worthless for any kind of research. I tend to agree with him. While I've found some interesting material at Wikipedia I couldn't honestly use it, given the anonymous nature of the thing. I've also seen errors go unchecked for months.I also find Morville's comments quite astonishing that he seems to think that authority derives from visual design among other things. 'Widespread faith in intellectual honesty'? Oh, come on now! Many (of course not all) entries are written by people with a particular axe to grind or a specific bias. One only has to look at the entries for Abortion to see that it is constantly being changed, with one bias being replaced by another.I do not know of a single librarian who will happily accept the information that they find it - and I speak to a *lot* of librarians.

Aha. I get it now. I didn't get that from the article, but putting it that way makes more sense.That being said... In many ways authority really is objective, and always will be. So what Wiki-Google-Tagged world is building might be a new kind of authority, one that's more fluid (Ambient perhaps?) and one that is built on many opinions from many people over time. This is just online though, does it transfer to offline stuff as well? Can we use this and apply it to our OPACs or maybe WORLDCAR?How can libraries use this is my big question.

I agree that authority is ascribed to a source by the consumer, and that peer-reviewed publication fails to ensure lack of bias. A very good example of this is one of my deans. His favourite font is MS Comic Sans, which I have seen him use for entire powerpoint presentations to the faculty when presenting very weighty information. While I know that the content is authoritative, the presentation does detract from the gravitas of the meaning. But the AUTHORity of the content is assured, regardless of the visual design.

And I also think that if I had to choose between a signed article by somebody recognized as an expert in the field and an anonymous, committee-created article, I'm more likely to choose the expert. Of course, experts have been known to say very silly things outside of their area of expertise (cf Clarke's first law [yes, I'm aware of the irony]).

In my new book, Ambient Findability, I make the following controversial point:

"Like relevance, authority is subjective and ascribed by the viewer."

I don't expect traditional librarians to immediately agree with this point. Heck, even Jimmy Wales (the founder of Wikipedia) disagrees with me. He believes authority is objective.

But I do expect that over time, more librarians will begin to realize that many of the traditional criteria for evaluating authority are just as superficial as visual design and information architecture.

Simply because something is published in the Wall Street Journal or a peer reviewed journal or written by a world-renowned expert does not assure quality or accuracy or lack of bias.

After my article on authority, Gene Smith conducted a follow-up tagsonomy interview. Enjoy!